Ion-Coupled Transport and Transporters
Chapter
74
PETER C. MALONEY and T. HASTINGS WILSON
This chapter concerns transport systems known as carriers, transporters, permeases, or facilitators. These are systems each requiring a single polypeptide that functions to catalyze substrate movement along an electrochemical gradient. Because chemical or photochemical energy transductions are not involved and because substrates simply move across boundaries, these systems are also said to mediate reactions of "secondary" transport. Such membrane carriers (transporters, facilitators, permeases) are associated with many different events in biology, ranging from the accumulation of nutrients and the extrusion of wastes by bacteria (discussed here) to the cycling of neurotransmitters at synaptic or synaptosomal membranes. But whether one considers microbiology or neuroscience, it is clear that these transport catalysts all share certain basic elements of structure and mechanism. This fact suggests that progress in understanding any single example might have an impact on understanding all others. Clearly, because bacterial systems are so tractable to genetic manipulation, a discussion of microbial transporters has special relevance to the larger field.
Several useful reviews drawing mostly on bacterial examples have appeared recently (70, 72, 97, 110, 142), so we are able to emphasize here topics not discussed elsewhere. In particular, there is an extensive account of a few important model systems and less of an all-encompassing discussion of examples found in Escherichia coli and Salmonella typhimurium (official designation, Salmonella enterica serovar Typhimurium).
This area of study was established in 1952, when Widdas (207) presented his analysis of glucose transport by the GLUT1 isoform of the mammalian glucose carrier (a relative of several hexose transporters in E. coli). Before this study, those interested in the transport of sugars, amino acids, or other organic substrates interpreted findings by using a framework developed to model ionic distributions across membranes. For this reason, data were typically evaluated in terms of coefficients of diffusion and permeability. This framework was especially productive in the study of ion transport but was less useful in understanding sugar or amino acid transport, for such parameters varied in unexpected ways, at times even showing a dependence on the internal concentration of transported solute (207, 208).
Widdas showed how such inconsistencies were accommodated if viewed from a new perspective. His "carrier" hypothesis (207) suggested that transport was initiated by binding of substrate to a carrier (Fig. 1, middle) acting as the catalytic element. Consequently, the kinetics of glucose transport would include terms related to substrate binding. He also proposed that the carrier-substrate complex could "diffuse" across the membrane, so that transport kinetics would also include terms describing the mobility or permeability of the complex, much as the older paradigm required (125). At the opposite surface, the carrier-substrate complex would dissociate, the unoccupied carrier would "return" to the original membrane surface, and the process could begin again. This new mechanistic view and simple extensions of it (Fig. 1) are as successful now as they were in the 1950s in describing the kinetics of carrier-mediated transport. Of course, it is unlikely that carriers diffuse across the membrane. Instead, we see the diffusion and permeability constants of the carrier and carrier-substrate complex as rate constants describing the alternate exposure of substrate binding centers to one or the other membrane surface. Accordingly, substrates are now seen as passing through rather than being carried by transport proteins.
The next step forward was taken only a year later, in 1953, when Mitchell and Moyle presented a study of Pi exchange in the bacterium Micrococcus pyogenes (now Staphylococcus aureus) (123, 126). This reaction is now assigned to a transporter resembling UhpT, one of several anion-exchange proteins found in E. coli and S. typhimurium (87, 111). The most significant aspect of this early work was its experimental proof that a membrane carrier acted as a catalyst. Specifically, Mitchell and Moyle (123, 126) showed that Pi exchange was sensitive to inhibition by mercury, and they documented that inhibition was achieved with far fewer mercuric ions than there were internal Pi molecules available for exchange (a ratio of about 1 in 30). Clearly, then, the mercury-sensitive element (a protein or protein-lipid complex) had to act catalytically. Work with a bacterial system also represented the third step forward in this field, for in 1956 Monod and his collaborators described the lactose carrier (LacY) in E. coli (158), bringing genetic tools to the study of membrane transport. LacY spent its early years in the shadow of its soluble sibling, β-galactosidase, but has now emerged as the best-studied example of a membrane transporter. The final formative event of the 1950s was the insight by Crane, in 1959, that a carrier might combine with more than one substrate at a time (Fig. 1) (39), thereby explaining the Na dependence of glucose transport by intestinal tissue. Following his own thoughts concerning the role of protons in membrane biology, Mitchell suggested in 1963 that the mechanistic base of LacY function was an H+/lactose cotransport reaction (124). Both ideas have proven to be correct.
The effect of these early studies was to establish a theoretical framework whose credentials were experimentally testable. Although these tests were not always easy to interpret, we now accept these ideas as the most accurate shorthand that describes carrier-mediated events. Indeed, whether we consider the capture of nutrients or the recycling of neurotransmitters, some combination or elaboration of these three prototypical systems—GLUT1, UhpT, and LacY—offers an adequate explanation of the phenomenon under study.
These early experiments also introduced model systems representing the three main biochemical mechanisms of carrier-mediated transport (Fig. 1). Thus, using the terminology of Mitchell, we describe glucose movement by GLUT1 as a reaction of "uniport," one in which a single substrate moves along its electrochemical gradient. The reaction mediated by UhpT exemplifies exchange, or "antiport," while LacY carries out the one-for-one cotransport or "symport" of a proton and a disaccharide, lactose. In this last case, note that tight "coupling" (125) of the cosubstrate fluxes requires that the binary complexes be unable to reorient (Fig. 1, right).
Although these reaction schemes (Fig. 1) have quite different kinetic consequences, we believe that a single protein class displays all three mechanisms. The evidence supporting this idea comes most importantly from study of the large number of carrier sequences now known in various prokaryote and eukaryote organisms. Analysis of this database confirms an early suspicion (106, 109), that most examples have hydrophobic cores containing 10 to 12 likely transmembrane α-helices (Fig. 2). This kind of structure was first suggested for the LacY protein (51), following Kyte and Doolittle’s suggestion (99) that transmembrane α-helices might be identified by screening an amino acid sequence for unusually long stretches of hydrophobic residues.
Several hundred carrier sequences have been examined (see below), and together they define two major groups. The schematic of Fig. 2 appears to represent most cases (perhaps 80% of the total), including examples of each reaction type (Fig. 1) and representatives from bacteria, fungi, and animal cells (71, 114, 157). The presence of this common structure most often arises from a theoretical analysis of the amino acid sequence, but in a growing number of cases, there is independent supporting evidence, particularly as provided by various reporter groups inserted into or fused with the target protein (see below). Those examples in the minority group, composed largely of examples from eukaryotic organelles (mitochondria, chloroplasts, or peroxisomes), are smaller and show a hydrophobic core of five to seven transmembrane α-helices (98, 130). Among these several hundred examples, however, only a few have also been the target of biochemical study sufficient to determine the minimal active unit. When this biochemical information is taken into account, however, a simplifying pattern emerges. Thus, transporters in the larger group appear to function as monomers (1, 26, 38, 104), while those of the smaller group work as dimers (7, 203). This hypothesis is made somewhat less testable than one would like because some eukaryotic carriers form tightly associated oligomers in vitro (73, 104), and special conditions are needed to assess the functionality of the monomer (26, 104). With this reservation in mind, however, one is justified in arguing that a single organizational principle characterizes all of these systems: that the smallest unit of biochemical function has a hydrophobic core of ca. 10 to 12 transmembrane α-helices (Fig. 2).
Evidence of an underlying unity among membrane carriers can also come from the detailed study of a single example. For example, King and Wilson (95) and Brooker and his colleagues (20, 21, 61, 116) found that certain LacY mutants have an uncoupled behavior in that the two cosubstrates (H+ and lactose) may be transported independently of one another. The simplest explanation of these observations is that in contrast to the wild-type protein, in which only the ternary complex (CSH) reorients, in these mutants there is reorientation of the binary complexes (CS or CH, respectively; Fig. 1, right). Other work reveals LacY variants that show a substrate exchange (antiport) reaction very much favored over net substrate transport (29), so that altogether, LacY validates the work of the 1950s by confirming that the simpler kinetic forms are embedded in the more complex reaction scheme (Fig. 1).
If membrane carriers share a common structure (Fig. 2), one might also expect an evolutionary relatedness among the various examples. Indeed, such relationships are well defined for a number of carriers in bacteria (62, 105, 114, 157) and also for carriers in their mitochondrial descendants (80, 98, 130). These studies establish two large and self-contained groups that encompass 105 cases from a total of 175 amino acid sequences examined (more are now available, of course). The more heterogeneous collection is the major facilitator superfamily (MFS) (114) (Fig. 3). As first described by Marger and Saier (114), the MFS contained 67 carriers distributed over five major clusters, with all examples showing 10 to 12 transmembrane helices (Fig. 2), whatever the reaction type (Fig. 1) or organism (both prokaryotic and eukaryotic carriers are in the MFS). Na+-coupled porters were not originally found in the MFS, but more recent work (157) now assigns six such sequences to a sixth cluster within the MFS (not shown in Fig. 3). The other large collection of related carriers contains the 38 known sequences encoding mitochondrial carriers as well as one peroxisomal example (80, 98, 130).
These extensive sequence alignments can test early speculations and reinforce or reject earlier observations. For example, we see no large-scale segregation of prokaryotic and eukaryotic examples within the MFS, and individual families and subfamilies are more likely to be defined by their substrates or reaction mechanisms, in line with the idea that such transporters were present even in the earliest cells (113). The database also verifies that the internal duplication apparent in certain tetracycline resistance proteins (168) is more broadly represented, particularly within clusters 1 and 2 of the MFS (62, 114). Thus, for these carriers, the N-terminal portion of the molecule (helices 1 to 6) often has significant homology with the C-terminal portion (helices 7 to 12). Such conclusions raise important issues for discussion and experiment (109), for they suggest that members of the MFS have arisen following gene duplication and gene fusion. Accordingly, the ancestral protein may have been a homodimer, and the contemporary form may be considered a covalent heterodimer. For these reasons, one expects elements of symmetry in the structure of MFS carriers, a prediction which may soon be tested (110).
Sequence comparisons among members of the MFS have revealed a strongly conserved motif that may serve as a signature for a number of these family members. Between transmembrane helices 2 and 3 as well as between their C-terminal equivalents, helices 8 and 9, one often finds an RXGRR motif (although K may substitute for the first and last R residues) (62, 114). In probing the significance of this conserved motif, Yamaguchi et al. (218) made various substitutions during site-directed mutagenesis of the Tn10 tetracycline resistance carrier (TetA), an antiporter that extrudes a (Mg-tetracycline)+ complex in exchange for H+ (216). For the region between helices 2 and 3, replacement of glycine (G69) by eight different residues gave TetA activity that was directly proportional to the ability of the residue to participate in a β turn. A positive charge at the next position (R70) appeared to be essential, but the peripheral arginines (R67, R71) were each dispensable. In LacY and several tetracycline carriers, there are, in addition to the RXGRR sequence, serine and aspartate residues preceding the first R. The anionic residue appears to be essential for TetA function, while serine does not (217). Such work suggests that the first of these motifs may play a structural role, but there is as yet no evidence that they take part in any specific aspect of membrane transport.
Representative Examples.
The examples summarized in Table 1 are cases in E. coli or S. typhimurium in which direct evidence or indirect tests (e.g., homology) are sufficient for a likely classification of the reaction catalyzed as uniport, symport, or antiport. By and large, the experimental findings are consistent with what one might expect. It is not surprising that bacteria move glycerol by uniport, since the accumulation of this rather permeant substrate via antiport or symport would be an unwise investment. Nor is it surprising that uniport is poorly represented. Microorganisms often scavenge or recapture nutrients from low external concentrations, and uniport, which allows only an electrochemical equilibrium, does not always fit this need. On the other hand, symport reactions are well suited to the net accumulation of material from the medium, meeting the demands of many amino acid and carbohydrate transport systems. Table 1 also includes examples of symporters that do not depend directly on H+ but use Na+ or Mg2+ instead. Such diversity should be taken for granted and is likely to be underestimated by the listing given here, because if the affinity for Na+ is high, one may easily mistake an Na+-coupled symporter for an H+-coupled one, as occurred during the study of PutP (185).
Table 1Chemiosmotic porters in E. coli and S. typhimuriuma |
Poise and Balance.
The listing in Table 1 suggests that if one searches diligently, a chemiosmotic transporter may be found to satisfy almost any physiological demand. If only for this reason, interactions among these systems are bound to be complex, and it becomes important to find organizing principles. The following expression (45, 108, 161) may be useful in this regard, since it describes the necessary thermodynamic limits. Thus (units are millivolts),
where it is understood that
In such cases, z 1 is the valence of the coupling ion (Ion), z 2 is the valence of the substrate (S), Δμ Ion is the electrochemical gradient for the coupling ion, ΔΨ is the membrane potential, and other terms have their conventional meanings (as in chapter 19). This formalism states only that the final distribution of substrate (if not metabolized), [S]in/[S]out, reflects a balance between its own tendency to distribute according to its electrochemical gradient [z 2 ΔΨ + (RT/F)...] and an opposing tendency determined by the coupling of substrate transport to some ionic circulation at stoichiometry, n. Thus, when n = 0 (uniport), substrate distributes passively; for a nonelectrolyte, then (e.g., glycerol, where z 2 = 0), internal levels of substrate would not rise above the medium concentration ([S]in/[S]out = 1). During antiport, n would take a negative value. In this case, also note that 
is itself a composite reflecting electrical and chemical forces acting on the coupling ion. Thus, when there is neutral exchange (e.g., Ca2+/2H+ antiport), the terms describing voltage will cancel (–nz 1 ΔΨ + z 2 ΔΨ), and only the respective chemical gradients will determine substrate distribution. Finally, when n takes on a positive value during symport, the extent to which substrate distribution is influenced by chemical and/or electrical gradients depends again on both the stoichiometry of coupling, n, and the valencies, z 1 and z 2.
This thermodynamic analysis has been important to the rational interpretation of events in both prokaryotic and eukaryotic cells. The first adequate study, by Kashket and Wilson (89), dealt with galactoside distributions in bacteria, and in that case, the collected data documented a coupling ratio of one H+ per sugar over a substantial range of the individual parameters. However, that instance was of fortunate simplicity: a nonelectrolyte substrate (z 2 = 0) and a single ionic species used in coupling (H+) at the simplest of stoichiometries (n = 1). A more realistic view is that such coupling is as involuted as biology demands, so the complete relationship noted here may be of value.
The simplicity of these mechanistic and structural descriptions (above) belies the complexity with which transporters integrate into cell physiology. One way to illustrate this is by reference to Fig. 4, which shows several porters at work in E. coli. Operation of these carrier systems is determined by the ambient ion motive gradient(s) (see chapter 19). As noted earlier, uniport (Fig. 4, reaction 1) is poorly represented in E. coli, perhaps because bacterial carriers mainly function to (re)capture substrates present at low external concentrations, an aim not well served by uniport unless the substrate carries a positive charge (but see the comment about FocA below). By contrast, H+-coupled symport (Fig. 4, reaction 2) is common in microorganisms, as substrate accumulation is driven by the electrical and chemical gradients established by an outwardly directed proton pump. LacY exemplifies this reaction, and the diagram in Fig. 4 correctly illustrates its normal function. On the other hand, suppose the galactose derived from lactose hydrolysis is not metabolized. Galactose would then accumulate internally, and if the carrier shows even low affinity for galactose, outward flux of the monosaccharide by way of the carrier would introduce a new element, the outwardly poised galactose gradient, contributing to lactose movement. Some time ago, this possibility was excluded for LacY (166), but more recent work shows this to be a realistic scenario in other cells (142), so the general possibility should be left open. A similarly elaborate situation occurs during operation of Pit, which catalyzes the symport of H+ with a neutral complex of metal and phosphate (e.g., H+/MgHPO4) (197). Transport of phosphate via Pit has all the characteristics of a proton symport, but because other systems regulate internal Mg2+, influx of Pi may be largely irreversible owing to lack of its metal cosubstrate for efflux.
Ion-exchange carriers likewise display unexpected properties at the level of physiology. For example, the usual view of Na+/H+ antiport (Fig. 4, reaction 3) has it serving two related functions. The reaction extrudes Na+ leaking inward by nonspecific routes, maintaining an inwardly directed Na+ chemical gradient. In addition, the exchange allows continued operation of carriers that bring Na+ inward by way of more specific pathways (Fig. 4, reaction 4). Here, too, tradition is overly simplistic. Na+ homeostasis is now being attacked at a molecular level (138, 175), and we have come to appreciate that Na+/H+ antiport is not a one-for-one affair but is weighted in favor of protons (138, 175, 190). We also know that E. coli has two Na+/H+ antiporters (NhaA and NhaB), each with its distinct and asymmetric stoichiometry (138, 175), and that one of these, NhaA, is strongly accelerated by alkaline pH (190). One result is that along with its participation as described above (Fig. 4), Na+ can play a role in the regulation of internal pH under abnormally alkaline conditions. Perhaps more interesting, with this new understanding of genes and gene products, one may use NhaA-NhaB null strains as vehicles for the cloning of new cation transporters with low affinities for Na+. It was in this way that Ivey et al. (78) identified a Ca2+/H+ antiporter known previously only from biochemical work (5, 162).
The physiology of anion exchange is also more complex than anticipated, and in E. coli this is nicely illustrated by the probable operation of UhpT (Fig. 4, reaction 5). Biochemical studies in Streptococcus lactis (now Lactococcus lactis) and Staphylococcus aureus suggest that UhpT catalyzes a neutral exchange involving the sequential inward and outward movements of two negative charges (111). Unexpectedly, UhpT accepts as its passengers either a pair of monovalent anions (e.g., 2HG6P1–) or a single divalent species (1G6P2–) (4). Therefore, as cells grow on glucose 6-phosphate (G6P), one envisions an unusual two-for-one antiport of monovalent and divalent G6P anions (Fig. 4) because of the relative acidity of the extracellular space. In this way, a reaction based mechanistically on anion exchange is transformed at the level of physiology into one that behaves like 2H+/G6P2– symport.
As a final illustration of the elaboration that comes with higher levels of organization, one might consider the likely operation of CadB, the lysine0/cadaverine1+ antiporter encoded within the E. coli cad operon (119) (Table 1). The coupling of lysine entry and cadaverine efflux would generate a membrane potential, with the inside negative, while decarboxylation transforming lysine into cadaverine consumes a cytosolic proton. Therefore, the overall sequence (lysine entry → lysine decarboxylation → cadaverine exit) acts as an outwardly directed proton pump, similar in design to a growing list of other proton motive metabolic cycles (see chapter 19). A similar principle may be at work during the exchange of ornithine and putrescine via PotE (Table 1), and in a somewhat less obvious way (167), the same thing may be said of formate1– entry via FocA. These and the preceding examples show convincingly that despite structural and biochemical uniformity, membrane transporters engage in an unexpected variety of events as biology emerges from biochemistry. At the very least, we should not be surprised if such elaborate case histories continue to appear.
In the remainder of this chapter, we deal mainly with E. coli and S. typhimurium carriers that serve as important model systems. Largely because these examples are amenable to molecular biology, we are more and more confident about structure-function relationships here, and even the near future promises significant advances. In the category of uniport, we discuss the transport of glycerol, which has certain unexpected features that make it of special interest. Of the antiporters in E. coli and S. typhimurium, we concentrate on the analysis of anion exchange, emphasizing UhpT and its relatives (111). In this last case, we comment specifically on the complex physiology associated with anion exchange and highlight recent findings concerning the pathway through this transporter (220, 220a). These latter findings may offer a new approach to understanding structure-function relations in membrane carriers. As an example of what can be learned from studies of symport, we next provide a summary of the LacY protein, the most exhaustively studied of membrane carriers. Because LacY is within the MFS, one expects this information to reinforce the study of others in this superfamily and vice versa; such cross-fertilization will be of great value, since it is now clear that no single system is likely to be accessible to all desired experimental approaches. We also consider melibiose transport via MelB, since this symporter shows an unexpected plasticity in its cation selectivity. The origin of this behavior is unknown, but because part of the cation-binding site of MelB has been localized, it may soon be possible to arrive at reasonable suggestions as to how these decisions are made.
During uniport, a substrate moves independently of other coupling elements (equation 1 and its accompanying text). This thermodynamic principle applies equally to an ion channel, but carriers and channels can be distinguished experimentally (184). In the case of glycerol transport, these tests have been difficult to conduct for technical reasons. But as it happens, transport of glycerol in E. coli may not be by a carrier after all.
The presence of a glycerol facilitator had been deduced by Sanno et al. (172) on the basis of cell volume measurements. For example, addition of high concentrations of glycerol (≥0.2 M) caused cells to shrink abruptly because of rapid loss of cell water. Cells with the glycerol facilitator rapidly returned to their normal volume as glycerol entered (and water followed), while cells lacking the facilitator showed a much reduced rate of swelling. The more detailed studies of Heller et al. (69) favor a pore-type mechanism for the glycerol (polyol) facilitator, as does a recent study of this facilitator expressed in Xenopus oocytes (117), and a similar conclusion is reached by examination of the amino acid sequence provided by Sweet et al. (189). This sequence indicates a hydrophobic protein with six transmembrane segments, itself an atypical finding for bacterial transporters. More striking, the facilitator (GlpF) shows homology to a number of eukaryotic membrane proteins, some of which are known to mediate water or urea movement as channels or pores structured as homotetramers. Collectively, these proteins belong to a group known as MIP, a term referring to the major intrinsic membrane protein of bovine lens, the first member of this group to be sequenced (156). The propanediol facilitator, PduF (Table 1), is also a member of this family (67% identity to GlpF).
All this points to GlpF as the first well-studied case of a channel in the E. coli inner membrane. An earlier argument (108) had suggested that E. coli would not have channel proteins, for if a cation channel of typical conductance were to open even transiently, the cell might risk a toxic depolarization or a lethal uncoupling. However, these risks are not present if one considers the movements of nonelectrolytes, so that use of a selective channel to accelerate glycerol (re)entry is entirely appropriate. Anion uniporters, or possibly "channels" (Table 1), are similarly "nontoxic" and are expected to be readily tolerated.
Background.
Among anion exchangers in bacteria, the Pi-linked examples are the best characterized, and Kadner et al. (87) and Maloney et al. (111) provide extensive reviews on the regulation and operation of these systems. Each accepts an organic phosphate as its primary (high-affinity) substrate, and each, as implied by the name, also accepts Pi, usually with a relatively low affinity. It is of interest that these systems accept both organic and inorganic substrates, but these antiporters are of significance to current science for other reasons. (i) They show an unusual ionic selectivity in accepting their substrates, leading to a pH-dependent variable exchange stoichiometry. (ii) The biochemistry of these exchangers integrates into cell physiology in an unexpected way (e.g., Fig. 4). (iii) Recent work holds promise of, quite literally, identifying amino acid residues that line the transport pathway through these proteins.
The first description of these reactions occurred in the 1950s, when Mitchell (123) and Mitchell and Moyle (126) found that resting cells of Staphylococcus aureus catalyzed a Pi exchange selecting the monovalent Pi anion. Ten years later, Harold et al. (67) arrived at a different conclusion for Streptococcus faecalis (now Enterococcus hirae), in which Pi transport depended on concurrent metabolism (ATP) and selected divalent Pi. A diversity among Pi transporters is even more broadly felt in E. coli, where we now document at least four reactions that carry 32Pi into the cell (10, 163, 164, 165, 209). Two of these (Pst and Pit) are specific for Pi itself, while the other two (GlpT and UhpT) accept Pi as an analog of some preferred organic phosphate substrate. It is now clear that Pst is an ATP-driven uptake system belonging to the multidomain ABC transporter superfamily (see chapter 76), while Pit is a proton-linked metal-phosphate symporter (197). These are the usual solutions to bacterial Pi transport, and Pst-like energetics accommodates the findings in E. hirae (67), while Pit-like behavior accounts for respiratory-driven Pi transport in E. coli (182), Micrococcus lysodeikticus (56), Paracoccus denitrificans (27), or Staphylococcus aureus (183). Of the additional Pi transport systems, GlpT was initially described by Lin et al. (68) as a glycerol 3-phosphate (G3P) porter, whereas UhpT was discovered as a G6P carrier by Winkler (212) and Pogell et al. (141). Early schemes considered GlpT and UhpT as added examples of nH+/anion symport, but we now know that these are Pi-linked antiporters (2, 111, 182). Thus, a current summary has Pi transporters coming in two varieties. There are those designed for net Pi movement (e.g., Pit and Pst), and there are those that mediate an exchange involving Pi and/or a phosphorylated substrate (e.g., GlpT or UhpT). Hindsight shows that Harold et al. (67) described the former (dedicated) category, while Mitchell and Moyle (123, 126) studied one the latter (exchange) class.
Diagnostic Features.
Pi-linked exchange is widely distributed (44, 214). Moreover, because these systems were first identified in gram-positive cells, these examples provide important information. In particular, documentation of ionic selectivity is most complete for Staphylococcus aureus or L. lactis, as is the determination of exchange stoichiometry, while the molecular study of exchange dominates the work in E. coli. Thus, models of Pi-linked antiport draw on several different systems, each according to its strength.
Most information derives from four systems in three cells: UhpT and GlpT, which are, respectively, the G6P and G3P carriers in E. coli; and the UhpT-like transporters of L. lactis and Staphylococcus aureus (Table 2). All share in the following. (i) Each mediates the self-exchange of Pi (Pi:Pi exchange) (2, 112, 182, 183), and work with Staphylococcus aureus (126) and L. lactis (112) shows that such exchange favors monovalent Pi. (ii) Each catalyzes the heterologous exchange of Pi with a sugar phosphate (G6P, G3P, etc.); in E. coli (182) and Staphylococcus aureus (183), one can exclude operation of symport as the mechanism of sugar phosphate transport. (iii) In L. lactis (3, 112), Staphylococcus aureus (126), and E. coli (UhpT) (182), AsO4 freely substitutes for Pi. (iv) In these same cases, maximal rates of Pi:Pi exchange are about fivefold faster than those of heterologous exchanges, and while no complete kinetic analysis is available, there is direct competition in L. lactis (4) and E. coli (GlpT) (68) between Pi and the organic phosphate substrate. (v) An additional kinetic phenomenon (of uncertain significance) is that optimal activity requires high ionic strength, whether in L. lactis (112), E. coli UhpT (48), or S. typhimurium PgtP (171); this does not reflect a cation-linked symport (112). (vi) The evidence favors an electroneutral event for both homologous and heterologous antiport (L. lactis, Staphylococcus aureus, E. coli UhpT, S. typhimurium PgtP) (2, 112, 182, 183, 198). (vii) Homologies among the sequenced examples suggest a common origin and therefore a common mechanism (47, 57, 59, 135). (viii) Finally, it is likely that the antibiotic fosfomycin, a toxic analog of phosphoenolpyruvate, enters the cell as a result of transport by Pi-linked exchange (88, 214).
Table 2Pi-linked anion-exchange proteinsa |
Pi-linked exchange systems differ in their choice of primary substrate (Table 2), but there are difficulties in evaluating the breadth of substrates for even well-characterized examples (see reference 111 for a detailed listing). For example, quantitative information from the older literature is often compromised by the use of media containing Pi, and especially in gram-negative cells, surface or periplasmic phosphatases, isomerases, and mutases may convert precursors into authentic substrates. Also, the presence of both homologous and heterologous exchange modes causes confusion, especially when one is dealing with substrates of widely differing affinities (183). One result of this is that the biochemical and biological phenotypes can differ in unexpected ways. Both phenotypes are "correct," of course, but for different reasons in different contexts.
Biochemical Mechanism: Inferences from Exchange Stoichiometry.
The biochemical model of Pi-linked exchange comes from knowledge of exchange stoichiometry, which for technical reasons is most confidently studied with the UhpT protein of L. lactis. Such work shows two Pi exchanging for one G6P in assays at pH 7 (3). Since the reaction is electrically neutral and since the homologous Pi:Pi exchange uses monovalent Pi, a two-for-one heterologous antiport likely reflects two monovalent Pi anions exchanging for the divalent G6P anion. This exchange is affected in two important ways by assay pH. First, acidification reduces overall function because of a decreased maximal velocity with little change in substrate affinity (4). Accordingly, the pH independence in affinity for G6P, over a pH range that spans its own pK2 (= 6.1), implies that both monovalent and divalent sugar phosphate anions are effective substrates. The second effect of pH is to change heterologous exchange stoichiometry directly: the antiport ratio (Pi:G6P) moves from an upper limit of 2:1 at pH 7 to a lower limit of 1:1 at pH 5.2.
These facts are organized by a simple model (4) whose consequences were discussed earlier (Fig. 4). In brief, the model proposes that UhpT has a bifunctional active site that accepts either two monovalent anions or a single divalent species but not both. (Think of A23187, an ionophore that uses two carboxylate anions to bind either a pair of H+ ions or a single Ca2+ [Mg2+] during a 2H:Ca neutral antiport.) This ensures an electroneutral exchange with simple ping-pong kinetics as the antiporter samples the cis and trans compartments. Given a preference for monovalent over divalent Pi, one can now understand the finding of a 2:1 (Pi:G6P) exchange stoichiometry at pH 7, where sugar phosphate is mostly divalent. The model also explains a pH-dependent change in stoichiometry, provided that both mono- and divalent sugar phosphates are acceptable (as argued). Thus, there would be two kinds of substrates: monovalent anions (monovalent Pi, sugar phosphate, etc.) exchanging at a ratio of 2:2; and divalent anions (typically, sugar phosphates) that move at 1:1. Macroscopic stoichiometry would then be a mixture of these elementary events weighted according to the relative abundance of each substrate type and the strength of its interaction with the carrier.
The details of this model depend on experiments with L. lactis, but the general idea allows one to understand findings elsewhere. For example, Essenberg and Kornberg (48) proposed an nH+/G6P symport mechanism for UhpT in E. coli, and much evidence points to this as the appropriate functional view (48, 66, 213). Indeed, one expects this, because the E. coli cytoplasm is more alkaline than the periplasmic space and thereby drives a cycle in which net G6P accumulation arises from the exchange of two monovalent G6P anions and a single divalent species (Fig. 4). Phenotypically, this mechanism appears as a symport with protons, avoiding the otherwise paradoxical conflict between thermodynamics (symport) and mechanism (antiport).
Probing the Translocation Pathway.
Tests of this model (Fig. 4) with traditional biochemical and biophysical techniques will be difficult, since these methods are poorly developed in membrane biology, and future work on Pi-linked exchange must emphasize other tools. Fortunately, the molecular credentials of this family are well established, largely as a result of efforts by Kadner, Boos, Hong, and their collaborators, who provided the sequences of Pi-linked antiporters in E. coli and S. typhimurium (47, 57, 59) and established that these carriers have a topology conforming to the 12-helix model discussed earlier (60, 77). As a result, the next generation of experiments will use molecular biology as a probe of structure and function. One important outcome of this is development of a line of experiments (below) that may localize the translocation pathway through UhpT and thereby describe the amino acid residues that interact with substrate. If this proves feasible, it should soon be possible to think more intelligently about questions of selectivity and stoichiometry.
In work with UhpT, Yan and Maloney (220) found a residue on transmembrane helix 7 that must be on the transport pathway. This conclusion arises from a study designed to identify mercury-reactive cysteine(s). By using site-directed mutagenesis to generate single-cysteine variants, two of the six cysteines in UhpT (C143 and C265) were identified as sensitive to the permeant SH-reactive probe p-mercuribenzoic acid. Further work with the impermeant derivative, p-mercuribenzosulfonate (PCMBS), showed C143 to be accessible only from the cytoplasmic phase, as expected from the likely topology of UhpT (77). By contrast, C265 was accessible to this impermeant and hydrophilic probe from both membrane surfaces (220); that is, inhibition of UhpT function (and its protection by substrate) was found for both intact cells and everted membrane vesicles.
That work (220) offers a simple experimental test identifying residues in the translocation pathway: their accessibility to an impermeant, hydrophilic probe from both membrane surfaces. With this in mind, now recall that membrane carriers operate to transport a single substrate (or a set of substrates) per turnover (Fig. 1). Clearly, then, the translocation pathway must have three domains (Fig. 5). Two of these may be termed "peripheral," inasmuch as they are exposed to either the inside or the outside aqueous phase but not both, while the third or "core" domain must lie between these peripheral regions and become alternately exposed to both surfaces as the carrier reorients to accept and discharge its substrate(s). The physical dimensions of these areas are as yet unknown for any carrier, but in principle, their sizes may vary considerably. As an example, the core domain might be as small as a single chemical group on a single residue or as large as a cluster of residues distributed over perhaps 6 of the 12 transmembrane segments. All this will presumably vary according to the size of the substrate and the kinds and numbers of ligands required for its binding.
Work along these lines should be highly productive for UhpT, because a functional cysteineless version of UhpT is available (220). Therefore, one should be able to implant cysteines at new locations and use PCMBS accessibility to localize the residues and helices that delimit the translocation pathway. Indeed, the use of such cysteine-scanning mutagenesis shows promising early results. For example, it appears that the core domain extends along transmembrane helix 7 for at least 12 residues, from C265 through I276 (220a). Moreover, the pattern of PCMBS accessibility in this region suggests an α-helical structure to transmembrane helix 7, so this core domain would occupy a 17-Å (1 Å = 0.1 nm) span in the middle of the membrane (220a). If this dimension is replicated on other helices, the size of the pathway through UhpT could, in fact, enclose the numbers and kinds of substrates suggested by biochemical modeling (above). More important, because this work uses a simple combination of biochemistry and mutagenesis, the general approach may be applied elsewhere.
Recent work (6, 110, 128, 142, 143, 167) shows that antiporters can take part in a metabolic cycle that generates a proton motive force. The first example of this "proton motive metabolic cycle" was found in Oxalobacter formigenes, an anaerobe that uses oxalate transport and decarboxylation to sustain membrane energetics (6). Here, metabolic energy arises because the exchange of external oxalate2– with internal formate1– polarizes the membrane (interior negative) by import of a single negative charge. In parallel, there is the stoichiometric (one-for-one) consumption of an internal proton as oxalate is transformed into formate by decarboxylation (–OOC-COO– + H+ → HCOO– + CO2). Accordingly, this cell combines physically separated scalar (decarboxylation) and vectorial (antiport) reactions to arrive at a thermodynamic proton pump; as noted earlier, the same end result arises from the lysine-cadaverine exchange cycle of E. coli. Note that in both cases, H+, as such, is not pumped. Instead, its elemental parts are manipulated so as to give the same result: one negative charge appears internally, and in parallel, one acidic group disappears from the inside.
The prototype in O. formigenes is but one of a growing list of similar examples (see chapter 19), the best understood of which are those involved in malate decarboxylation (143), the conversion of histidine to histamine (by decarboxylation) (128), and the production of alanine by decarboxylation of aspartate (K. Abe and P. C. Maloney, unpublished data). As noted earlier, such examples are significant not so much for what they tell us about decarboxylating systems but for their illustration of how unexpected phenotypes arise when biochemically simple elements are arranged in a new way. The lesson is that we cannot be too complacent as we catalog more and more of microbial biochemistry and physiology (as in this book). Instead, we must continually be on the lookout for unexpected combinations of vectorial and scalar reaction ensembles (167).
Discovery.
An inducible lactose transport system in E. coli was first described in 1955 in a preliminary note from Monod’s laboratory at the Pasteur Institute (35), with the full report by Rickenberg et al. (158) appearing the following year. The term "permease" was coined to describe these systems (34), but the more recent literature has emphasized the terminology introduced by Mitchell (124, 125).
Substrates for the lactose transport system, LacY, include many β-galactosides and several α-galactosides (indeed, Pardee [140] independently discovered LacY by studying transport of melibiose, an α-galactoside), and under the usual circumstances, these substrates are metabolized sufficiently rapidly to avoid their accumulation within the cell. However, if β-galactosidase is removed by mutation, lactose or other β-galactosides will be taken up until they have accumulated to high internal concentrations, and one may characterize such "active" transport without the added complication of substrate metabolism. The same goal may be realized by using nonmetabolizable thiogalactosides, such as thiomethylgalactoside (TMG); these, too, will accumulate, typically to an internal level ≥100-fold higher than that of the external medium.
In 1963, Mitchell (124) proposed that sugar accumulation mediated by LacY was best understood as reflecting an H+/lactose cotransport reaction. This hypothesis remained a speculation until 1970, when West (205) showed that addition of substrate to energy-depleted cells resulted in proton uptake and alkalinization of the external medium. A 1:1 stoichiometry for H+/galactoside symport was demonstrated in later quantitative work, where it was also established that proton entry was accompanied by entry of a single positive charge (206). If there is a required coupling between H+ and lactose, then an imposed proton motive force should drive sugar uptake into energy-depleted cells, and the magnitude of this response should be predicted from the size of the imposed driving force (equation 1). Quantitative support for this type of relationship has come from several laboratories, using intact cells, right-side-out membrane, or everted vesicles (49, 50, 89, 100, 153, 154). Taken together with the initial work of West and Mitchell (205, 206), these provided compelling evidence favoring the idea that LacY functions as an H+/lactose symport mechanism.
Identification, Reconstitution, and Purification of LacY.
A major advance in the biochemical characterization of LacY, and of membrane proteins in general, occurred when Fox et al. (52, 53, 91) devised a simple method for labeling the LacY protein with the sulfhydryl reactive agent N-ethylmaleimide (NEM). Initially, the active site of the carrier was protected with excess substrate, and nonradioactive NEM was added to mask nonspecific reactive sites on this and other proteins. Cells were then washed free of residual NEM and the protecting substrate, after which labeled NEM was added to bind to -SH groups which had been specifically protected by substrate. Despite the low signal-to-noise ratio of this maneuver, it was possible to specifically identify the lactose carrier (giving impetus to its continued study), to demonstrate that it was a membrane protein, and even to determine an approximate mass of 31,000 Da (81).
A second important step in the biochemical study of LacY centered on development of a method for solubilization of the protein so that its activity could be reconstituted into liposomes. In 1980, 15 years after the work of Fox and Kennedy (53), functional reconstitution was achieved by Newman and Wilson (133), who extracted membrane vesicles with an octylglucoside-phospholipid mixture. This extraction was followed by a dilution step to disperse the detergent and generate unilamellar proteoliposomes containing membrane proteins in the lipid bilayer. When these proteoliposomes were prepared from vesicles containing LacY, imposition of a membrane potential (negative inside) led to an immediate accumulation of labeled sugar. If proteoliposomes were made so as to contain internal unlabeled sugar, addition of external labeled sugar led to a pronounced accumulation of substrate due to the rapid exchange of external (labeled) sugar with the internal (unlabeled) material. This test proved to be an especially sensitive assay with which to monitor purification of the protein.
Several other methodological advances were necessary to the final purification of LacY. An essential step involved the cloning of lacY onto a high-copy-number plasmid by Teather et al. (191), permitting a 10-fold elevated expression over that found for fully induced cells. Other major steps concerned assays of function. One of these, with a photoactivated ligand, 4-nitro[2-3H]phenyl-α-d-galactopyranoside (86), was based on substrate binding, making it possible to trace the progress of LacY purification by "doping" the preparation with a small amount of covalently labeled material. The second important assay exploited reconstitution of activity in proteoliposomes by using the exchange test noted above, ensuring that a purification scheme could be devised to yield functional protein. With use of the overproducing strain and with these two assays in hand, Newman et al. (132) purified the carrier to homogeneity in functional form, setting the stage for a formal demonstration that LacY alone was sufficient for all aspects of sugar transport found in the intact cell (132, 199).
Structure and Topology.
Speculation concerning LacY structure was possible as soon as Buchel et al. (25) provided the sequence of the gene. The deduced amino acid sequence revealed 417 residues, 70% of which were hydrophobic, as might be expected of a membrane protein. This sequence was also highly basic, an unexpected finding we now rationalize as an expression of von Heijne’s "positive-inside" rule (202). Circular dichroism measurements showed that LacY is more than 80% helical in conformation, and this observation, along with an analysis of hydropathy, led Foster et al. (51) to propose that LacY contained a cytoplasmic N-terminal hydrophilic segment followed by 12 transmembrane α-helices and a hydrophilic C-terminal segment (e.g., Fig. 2). Strong general support for this model has come from a variety of studies, including Raman spectroscopy (201), immunological analysis (30, 32, 33, 41, 74, 75, 176, 177, 178), protease accessibility (58, 186), and susceptibility to chemical modifications (139).
While the model of Foster et al. (51) is probably correct in most respects, an important modification was required to account for the evidence, collected by King et al. (92), of a physical interaction between D237 and K358 (see below). This modification placed a charged residue (D237) in transmembrane helix 7. At first glance, this may seem inappropriate, but it is likely that the anionic nature of D237 is compensated for by its interaction with a previously unaccompanied cationic residue, K358. There is, in fact, good evidence that these two residues interact with each other, probably in a salt bridge.
Work using PhoA (alkaline phosphatase) as a reporter of topology provides strong and independent confirmation that LacY has 12 transmembrane segments (helices). Underlying such work is the assumption that if PhoA, devoid of its normal signal sequence, is fused to the C terminus of some target protein, the observed phosphatase activity will reflect the position of the fusion point. Thus, high activity will be found if target protein topology places the fusion point in the periplasm, where dimerization of alkaline phosphatase may occur; low activity is observed if the fusion point is in the cytoplasm, where functional phosphatase cannot form. Calamia and Manoil (28) have analyzed a series of 36 LacY-PhoA fusion proteins in which the point of fusion is distributed broadly within LacY. The distribution of PhoA activity in these fusions clearly predicts the presence of 12 transmembrane segments (Fig. 6).
Transport by LacY Deletion Variants.
Stochaj et al. (187, 188) showed that truncated LacY polypeptides containing residues 1 through 50, 1 through 143, or 1 through 174 are stably integrated into the lipid bilayer, so it appears that even a short segment of the LacY N-terminal region can fold and insert independently of other parts of the carrier, possibly serving to guide the folding of the remainder. Building on these observations, Bibi et al. (12) constructed a series of internal deletions so as to remove both even- and odd-numbered helices. These variants were inserted into the membrane (note that those with an odd number of helices may have their N and C termini on opposite membrane surfaces), and the study of their transport properties led to an important observation. Although unable to accumulate substrate, these deletion variants nevertheless retained sugar recognition and were able to facilitate lactose movement in a "downhill" direction as if uniport function were present. One mutant is particularly notable, since it has 80% of the normal downhill transport activity yet lacks transmembrane helices 2, 3, 4, and 5. This suggests that most of the LacY translocation pathway resides in a C-terminal domain containing helices 6 through 12.
Do Protons Move by a Charge Relay?
Much evidence suggests that H322, E325, and R302 are important to H+/lactose cotransport. For example, Padan et al. (137, 151) showed that H322R mutants do not accumulate sugar or transport H+ but that they do carry out downhill movement of lactose, and Puttner et al. described similar behavior for the H322N and Q variants (151, 152). In the same way, the E325A mutant catalyzes downhill lactose influx at high substrate concentration but without cotranslocation of H+. Mutants at position 302 show these properties as well, so that the R302L derivative has defects in net sugar accumulation and in sugar-sugar exchange, yet downhill lactose entry (without H+) is observed. Collectively, such data suggest that the normal functions of these three residues (R302, H322, and E325) are essential to proper coupling between H+ and lactose fluxes. Since these residues are probably close to one another (84, 85), they may act on the same step in coupling.
In chymotrypsin and other serine proteases, His, Asp, and Ser residues are known to act as components of a charge relay system (13, 14) essential to catalysis. In an imaginative argument Carrasco et al. (29) postulated that H322 and E325 might serve an analogous function for LacY. Thus, H+ might move, sequentially, from E325 to H322 to R302, thereby accounting for the phenotypes observed for the various mutants described above. Despite the appeal of this idea, which was consistent with the observations then available (120), subsequent findings suggest that the mechanism of H+ translocation is more complex. For example, Blow (13) points out that there is invariably a serine in the protease charge relays and that the anionic residue is always an aspartate, never a glutamate. In addition, continued study of H322 mutants casts doubt on the obligatory role of histidine in proton translocation. Replacement of H322 by tyrosine or phenylalanine yields carriers that mediate proton cotransport with lactose and melibiose, albeit with reduced affinity for sugars (93, 94). Moreover, Franco and Brooker (54) found that the H322N variant does have lactose-induced proton entry, and Brooker showed that the double mutant A177V-H322N has an H+/lactose stoichiometry close to 1 (20). At the very least, one may conclude that an imidazole moiety at position 322 is not an absolute requirement for H+ cotransport in LacY; this fact makes the charge relay mechanism as originally conceived less attractive than before.
Mutants with Altered Substrate Recognition.
To delineate the region of LacY involved in recognition of substrates, several groups have isolated mutants with altered substrate specificities. For example, building on the observation of Shuman and Beckwith (181) that a maltose-positive phenotype could arise by mutation in LacY, Brooker et al. (22, 23) isolated a number of LacY variants that recognize maltose: A177 → V or T, and Y236 → F, N, S, or H. All of these mutants (save Y236F and Y236N) were later found to accept sucrose (95). Further work (37, 55, 61, 115) implicates other positions as important to sugar recognition (T266, I303, S306, K319, and H322), a view strengthened by the finding that combinations of mutations at these positions can yield carriers that transport a trisaccharide (maltotriose) or a tetrasaccharide (136).
When these and positions with similar properties are placed on the LacY topological map, an unexpected consensus emerges (Fig. 7). Note that all of these positions are expected to lie in transmembrane helices, suggesting that substrate recognition occurs within the hydrophobic core of the protein. Note as well that when two or more of these positions lie on the same helix, they are in a registration that suggests that only a single face on any one helix is involved in substrate recognition. Finally, because these positions are clustered in neighboring helices (helices 5 through 10), one is tempted to group these elements together to form a single binding pocket. More direct tests are needed before this arrangement is acceptable as a physical model, but even now, these findings add a valuable third dimension to the standard arrangement of LacY.
Mutants That Leak.
Normally, coupling between H+ and lactose is absolute, and one cannot demonstrate net transport of either substrate without the other. In certain LacY mutants, however, this coupling is significantly affected. Transport of H+ in the absence of sugar was first observed in the A177V derivative studied by King and Wilson (95). In normal cells, proton extrusion by the respiratory chain establishes an inwardly directed proton motive force of near 150 mV (see chapter 19). In the A177V mutant, however, protons move back into the cell so rapidly (via LacY) that the proton motive force is reduced from 140 to about 70 mV (95). An important additional finding was that a β-galactoside substrate, thiodigalactoside, blocked this proton "leak," restoring the proton motive force to its normal value (95). This behavior suggests that the A177V mutant works as a H+ uniporter, i.e., that the binary complex CH can reorient (Fig. 1). One presumes that by driving formation of the CS and CHS forms (Fig. 1), sugar binding prevents this uniport function by lowering the steady-state concentration of CH.
Brooker (21) and Lee et al. (101) described another kind of H+ leak in the double mutant A177V-K319N (20, 21) and in the double and single mutants D240A-G268V and D240A (101). In these cases, addition of sugar causes a greatly excessive proton movement into cells ( >>1 H+ per sugar). Here it is likely that sugar enters with protons, that protons are discharged to the inside, and that sugar returns without protons, leading to net proton movement without requiring sugar accumulation (21). In the terminology of Fig. 1, uniport for H+ would arise by an internal cycle involving the CS and CSH complexes.
Role of Charged Residues.
Four anionic residues and four positive residues are currently placed in transmembrane α-helices of LacY. Since it is energetically unfavorable for there to be uncompensated charges in a hydrophobic environment, it is possible that several (or all) of these charged amino acids are paired with residues of opposite charge. For example, because of their proximity, the pairing of H322 and E325 had been suggested (29), as outlined earlier. King et al. (92) were led to suggest an interaction between the more distantly located residues D237 and K358 by a series of genetic experiments initiated in a search for revertants of transport-negative mutants. One such negative mutant had been identified as a K358T variant (92), while a second (from the collection of Muller-Hill) proved to be D237N. When melibiose-positive revertants of K358T were obtained, all were second-site mutations in which D237 had been replaced by a neutral residue, either asparagine (eight cases), tyrosine (two cases) or glycine (one case). On the other hand, melibiose-positive revertants of D237N showed a second-site mutation as K358Q. Thus, loss of a single charge gave a defective protein (K358T or D237N), while the additional loss of an opposite charge restored function (K348T-D237N or D237N-K358Q), strongly suggesting the presence of a salt bridge between D237 and K358 in the wild-type protein. Sahin-Toth et al. (169, 170) provided additional support for the idea by showing that replacement of either residue by cysteine causes loss of function, while replacement of both with cysteine yields an active protein. Further, near-normal function was retained when these two charges were exchanged (i.e., the D237K-K358D variant).
Although there may be a salt bridge between D237 and K358, it is clear that neither these residues nor their precise positioning is essential to sugar recognition or proton transport, for the double mutants can retain considerable activity. For these reasons, it is not possible to attribute a specific functional role to this postulated salt bridge, and it may be as suggested by Dunten et al. (46) that this salt bridge is important for protein folding or stability. At the least, we may conclude that these two residues directly interact with one another.
Evidence for a more complex interaction between charged residues is given by Lee et al. (101, 102). Thus, when transport-negative variants were generated by replacing K319 with leucine or asparagine, melibiose-positive second-site revertants showed loss of a negative charge either at D240 (replaced by G or V) or at E269 (replaced by Q). Among these, the more robust activity was seen when both D240 and K319 were altered, a finding verified by Sahin-Toth et al. (169), who found that the double mutant D240C-K319C showed significant activity. Together, these findings suggest that K319 may form a salt bridge with both D240 and E269. Jung et al. (82, 83) used site-directed fluorescence labeling to study these possible interactions further.
In thinking about a role for this more complex set of interactions, it may be relevant that these residues lie in the region implicated as important to sugar recognition (Fig. 7). Thus, D240 lies in helix 7, E269 lies in helix 8, and K319 lies in helix 10. Accordingly, this salt bridge may work in a direct way to influence LacY function, perhaps as described in Fig. 8. According to this view, sugar binds to a site accessible from the external phase, but net translocation is obstructed by the interaction of E269 and K319. The inwardly directed proton gradient would foster protonation of the E269 carboxyl, breaking the salt bridge and allowing a secondary interaction between K319 and D240, which opens a pathway to the inside, thus allowing net influx of sugar. Subsequent release of H+ from E269 (into the cytoplasm) would return the salt bridge to its original state, restoring an outwardly oriented sugar-binding site and giving the net inward movement of one H+ and one sugar. Whether this kind of model has value will now depend on work that identifies the substrate translocation pathway, possibly by application of the techniques discussed earlier in relation to probing the pathway through UhpT.
Site-Directed Mutations in the Lactose Carrier.
The LacY carrier has been the target of an extensive series of experiments based on site-directed mutagenesis, particularly in the laboratory of Kaback (reviewed by Roepe et al. in reference 159), and perhaps as many as 200 mutants have been constructed and screened for function. One important object of this work has been the systematic study of individual LacY residues, e.g., the 12 prolines (159), the 14 tyrosines (160), the 6 tryptophans (118), and the 8 cysteines (see below), any of which is a potential H+ carrier. No one of the LacY prolines has proved to be essential, but among Tyr→Phe variants, two (Y26F and Y336F), showed complete loss of function, and a third (Y236F) showed defects in net movements but not exchange; a defect in this last case is understandable, since position 236 is also involved in substrate recognition (Fig. 6). Other work has replaced all six native tryptophan residues with phenylalanine, again with retention of near-normal activity (118). As the recipient of new tryptophan residues, this tryptophanless mutant should prove useful for locating sites where changes of solvent exposure might be reported by "site-specific" fluorescence (82, 118).
Interest in the role of cysteine was a natural result of the early finding by Fox and Kennedy (53) that substrates protect LacY against inactivation by NEM. Morever, it was feasible that cysteine participated in proton translocation. An attack on these issues began with identification by Beyreuther et al. (11) of C148 as the substrate-protectable cysteine. Trumble et al. (192) and Viitanen et al. (200) then obtained the C148G mutant, and later work generated the C148S variant (131, 174), both of which had normal or near-normal activity. Clearly, while C148 may be near the substrate binding site, it plays no direct role in substrate recognition or transport. (Yamato and Anraku [219] arrived at similar conclusions about the two NEM-accessible and substrate-protectable cysteines of PutP, the Na+/proline symporter [129, 215].) Glycine, valine, and serine replacements of C154 were studied next (122); C154G is inactive, but C154S and C154V have ca. 20% wild-type activity. Brooker and Wilson (24) derived the C176S and C234S mutants, and Menick et al. (121) generated the remaining serine substitutions (C117S, C333S, C353S, and C355S). These show normal activity, leading to the broad conclusion that no single LacY cysteine is essential for function. With this information in hand, it was practical to proceed in the construction of a cysteineless derivative, one in which the initial rate of lactose transport is about one-third normal and the steady-state level of substrate accumulation is about half that usually found (79). These findings make it unlikely that sulfhydryl-disulfide interconversion plays a direct role in lactose transport, as suggested by Konings and Robillard (96). More important, the cysteineless LacY now provides a suitable host for the introduction of new cysteines to serve as targets of SH-directed probes (82, 83).
Other H+-Linked Symporters.
LacY is the best-studied membrane carrier in bacteria, but several other proton-linked sugar cotransport systems have been described (71, 72) (Table 1). Among the more informative of these are the arabinose (AraE) (106, 107), galactose (GalP) (72), and xylose (XylE) (42) porters, all of which lie within cluster 2 of the MFS. These are now the target of increasing biochemical work (71), in part because one may apply to them approaches used earlier to study the related GLUT1, including use of common inhibitors such as cytochalasin B (AraE, GalP) or forskolin (GalP) (204). The continued study of these porters should usefully supplement the study of LacY (and vice versa), for one suspects that they share not only a common mechanism of proton translocation but also certain features of ligand binding.
Introduction.
The melibiose carrier of E. coli (MelB) was discovered by Prestige and Pardee in 1965 (150), but the features that make it of special interest were not appreciated until a decade later, when it was shown to be a symporter with unusually broad cation specificity, accepting H+, Na+, or Li+ as coupling ions (8, 193). The cotransport of these ions is most clearly identified as described by Tsuchiya et al. (193, 194, 195, 196), using ion-specific electrodes to measure ion movement associated with sugar entry into de-energized cells. Those tests show that in the absence of Na+ and Li+ (<20 to 50 μM), melibiose enters with H+. However, Na+ (or Li+) readily competes with and displaces H+, so that Na+ is the likely physiological substrate for melibiose cotransport (Table 3). Even more unusual, these tests document that cation preference differs according to the accompanying sugar: α-galactosides utilize H+, Na+, or Li+, whereas β-galactosides such as lactose or the nonmetabolizable TMG use primarily Na+ or Li+ (231) (Table 3). Such observations implicate a functional interaction (direct or allosteric) between the sugar and cation-binding sites, a view supported by the finding that mutants of sugar specificity often have altered cation selectivity (see below).
Table 3Cation selectivity of melibiose transporters of E. coli, S. typhimurium, and K. pneumoniae |
That these different cations (Table 3) interact at a common site on MelB is directly supported by studies of substrate binding with a high-affinity ligand, p-nitrophenyl-α-d-galactopyranoside (α-NPG) (36, 40, 146), as well as by assays of transport (210). Thus, the binding of α-NPG, which shows an expected competitive inhibition by melibiose or TMG, is strongly enhanced by Na+ or Li+ because of an increased affinity, and a kinetic analysis of binding indicates that the three coupling cations (H+, Na+, and Li+) compete with each other (at the same site) and that binding likely occurs in a 1:1 stoichiometry with sugar, in agreement with the stoichiometry found during transport (9).
H3O+ versus H+ as the Translocated Species.
The finding that different cations interact with the coupling site of MelB has important implications. Boyer (18) points out that if a carrier can use both H+ and Na+ (or Li+), the protein-ion interactions are likely to be similar in each case. For this reason, he suggested that the proton is probably taken in the form of its hydrate (e.g., H3O+) rather than H+, making it less likely that proton translocation would occur via a chain of protonatable-deprotonatable groups, as in a "proton wire." Instead, H3O+ would more likely interact through coordination to electronegative oxygen and/or nitrogen atoms, so the mechanism of its binding and translocation would resemble that of any other solute. In principle, of course, H+ could show either or both interactions: H3O+ binding to a variety of amino acid residues followed by transfer of H+ from this complex to some protonatable group (recall the argument made for salt bridge function).
In MelB, the observed competition between H+ and Na+ (see above and reference 210) is consistent with their sharing a common pathway for transport, so there is now much interest in studies that correlate specific residues and regions with ion translocation (see below). It is also of interest that evolution from the proton energetics of early cells to the sodium economy of higher eukaryotes is understandable if Na+ and H+ use similar binding and transport mechanisms (113).
Structure of the Carrier.
Analysis of the DNA sequence of melB, as determined by Yazyu et al. (222), predicts that MelB is a hydrophobic protein with 12 transmembrane segments, and construction of MelB-PhoA fusions provides data consistent with this model (15). An immunological study (17) also shows that an antibody raised against the C-terminal decapeptide reacts with inside-out membrane vesicles, indicating that both N and C termini lie in the cytoplasm. The MelB protein in E. coli shows an 85% sequence identity to the melibiose carrier of S. typhimurium (127), a 78% sequence identity to a similar carrier in Klebsiella pneumoniae (63), and homology to the lactose carriers of Streptococcus thermophilus (144) and Lactobacillus bulbaricus (103). As noted below, these homologies have proven useful in the mapping of an Na+-binding site.
Mutations That Alter Cation Selectivity or Sugar Specificity.
Both genetic selection techniques and site-directed mutagenesis have yielded MelB mutants altered in cation selectivity or sugar specificity. Noting that lithium inhibited growth on melibiose, early experimenters isolated lithium-resistant mutants mapping to MelB. Niiya et al. (134) characterized one such mutant as a single-point mutation that changed P122 to serine. More striking, in addition to being lithium resistant, this mutant could no longer use H+ as a coupling cation (221); the properties of other Li+-resistant mutants have been described by the Tsuchiya laboratory (90, 179, 180).
Mutants altered in sugar recognition were first isolated by Botfield and Wilson (16) after cells were grown on melibiose in the presence of a nonmetabolizable inhibitor, TMG. The TMG resistance phenotype arose in a complex fashion from any of 23 unique amino acid substitutions mapping to 17 different positions in four distinct regions (Fig. 9). An unexpected observation was that for most of these sugar recognition mutants, there was also an altered cation selectivity. In particular, Li+ produced less inhibition of growth on melibiose than was found with the parental protein. Indeed, one mutant was the P122S variant isolated earlier on the basis of its lithium resistance (see above). The simultaneous change of MelB interaction with sugar and cations suggests an intimate relationship between the cation and sugar recognition sites, and this has led to a proposal (16) that one or more electronegative oxygen atoms on the sugar itself might provide a cation coordination center.
A number of MelB site-directed mutants have been characterized, and in some of these, there appears to be a primary effect on sugar recognition or transport kinetics rather than on cation recognition. Thus, Sarkar and Kaback (see reference 85) changed D361 to glycine or aspartate and observed a much reduced transport rate; this and the similar behavior of the D361A mutant are attributable to a reduced maximal velocity of transport (145). In addition, no effect on cation-sugar coupling was observed when serine was substituted for C106, C231, and C360 (M. C. Botfield, Ph.D. thesis, Harvard University, Cambridge, Mass., 1989) or when the seven MelB histidines were changed to arginine (and for H94, also to N or Q) (147, 148).
There are four aspartate residues in membrane-spanning regions of MelB, and attention has naturally focused on their possible role in cation recognition. D31 is in helix 1, D51 and D55 are in helix 2, and D120 lies in helix 4 (Fig. 10). Each of these aspartates has been replaced, individually, by cysteine or asparagine (149, 223), and for all save D31N there is loss of TMG transport; for D31N, however, TMG transport and MelB expression both fall to ca. 10% of normal, indicating a normal specific activity. To determine whether sugar recognition is altered in these mutants, direct binding assays have been carried out with α-NPG (149). Those experiments show the D31N variant behaving as the wild type does in respect to the Na+ stimulation of α-NPG binding; for the other mutants, however, while there is normal α-NPG binding supported by H+ (e.g., in the absence of Na+), the affinity increase usually caused by sodium salts is absent. Such studies suggest that D51, D55, and D120 play some role in cation recognition, while D31 does not.
A somewhat different approach has been used to examine the properties required at positions 51 and 120. In this case, an amber codon was introduced at each position, and plasmids encoding each mutant were placed into each of eight different amber suppressor hosts, allowing substitution of S, Q, Y, A, H, K, L, and E for either D51 or D120 (211). Since no MelB function was observed for any except the D → E substitutions, one may reasonably conclude that a negative charge is required at positions 51 and 120. Considered in the light of earlier work (see above), this finding leads one to question whether MelB variants containing glutamate at these positions might show unusual coupling properties.
Several workers have studied MelB variants in which glutamate has been substituted for aspartate at position 51, 55, or 120, and of these, the D51E and D120E variants are most interesting. For example, Pourcher et al. (149) found that the D51E mutant shows H+/melibiose cotransport and that this is not stimulated by Na+ or Li+ (much like the D51N/C variants noted earlier). D120E, on the other hand, cannot use H+ for cation/melibiose cotransport yet retains normal responses to Na+ or Li+. These findings are confirmed by tests of sugar-stimulated cation transport into anaerobically incubated cells: the D51E mutant shows near-normal H+ uptake, while the D120E variant gives no H+ uptake whatsoever (211). These mutants also show altered interactions with β-galactosides, being defective in the accumulation of both TMG (149) and lactose (211), despite the presence of Na+ or Li+.
The Melibiose Carrier of K. pneumoniae.
The melibiose carrier of K. pneumoniae shows sugar cotransport with H+ but not Na+ (63) (Table 3). Since the amino acid sequence of this protein is 75% identical to that of MelB in E. coli (222), it seemed possible that a residue(s) responsible for Na+ coupling in E. coli might be identified. The first approach involved construction of chimeras between the two carriers (64). Thus, the N-terminal two helices of the E. coli protein were fused with the C-terminal 10 helices of K. pneumoniae, generating an "E2/K10" carrier; other chimeric combinations included E4/K8, E6/K6, E8/K4, and E10/K2. Na+ stimulated melibiose transport in each of these constructs, suggesting that information for Na+ coupling resided within the first two helices of the E. coli MelB protein, in a 77-residue segment where the E. coli and Klebsiella proteins differ by only 5 amino acids. For this reason, the second phase of study relied on site- directed mutagenesis. Starting with the Klebsiella carrier, each of these five residues was changed, individually, to the amino acid found in the E. coli sequence (65). In four cases, the Klebsiella carrier showed wild-type behavior (failure to recognize Na+), but in the fifth case, particularly dramatic results were obtained. Here, as a result of the replacement of alanine by asparagine at position 58, the Klebsiella protein acquired both a pronounced Na+ stimulation of melibiose uptake and, reciprocally, a melibiose stimulation of Na+ transport. It is clearly of interest that the Na+-responsive residue in K. pneumoniae (A58N) corresponds to N54 in E. coli MelB and that this lies in a region where independent evidence (see above) suggests the presence of an Na+-responsive element. Whether this region is involved directly in cation binding and/or coupling cannot yet be determined, but much evidence now points to this small part of MelB (D51, N54, and D55) as contributing to a site of Na+ interaction.
Because the kinetic structures summarizing the carrier hypothesis (Fig. 1) can be mapped onto a single structure (Fig. 2), the current emphasis is on finding model systems that give a close look at this molecule in a few specific cases. In this effort, the models described here may be among the first to offer a new level of resolution, and the next few years should prove especially exciting as we move to describe how substrates interact with and move through transporting proteins. Recall that there appear to be two kinds of carriers: those that operate as dimers and have a helix number of 5 or 6 and those that function as monomers with a helix number of 10 to 12 (110). Recall as well that sequence comparisons (at least within parts of the MFS) suggest that the monomers might be considered covalent heterodimers, with related N- and C-terminal domains, each having five or six helices. Considering this, one might reasonably argue that a substrate(s) would move through a transporter at its dimer interface or the equivalent, the interface between N- and C-terminal domains of the covalent heterodimer (110). No relevant direct information is presently available to test such ideas, but this circumstance will not last long. On the one hand, direct visualization of a pathway could soon come from crystallography, which is most advanced for an erythrocyte antiport protein, band 3 (155). Alternatively, it should be feasible to combine biochemistry and molecular biology to follow clues provided by the bacterial examples discussed here. In UhpT, for example, several residues in the pathway are already known (220, 220a). In LacY or MelB, residues affecting substrate specificity suggest a starting point for a similar analysis of sidedness. In LacY especially, the clever use of protein design principles is making it possible to probe structure with single-residue resolution (82, 83). If these proposals survive such direct tests, the field will have attained a significant new plateau. At the moment, we arrive at a two-dimensional perspective with reasonable confidence. Understanding where substrates bind and move, even in only a few cases, could bring a similarly rational basis to inferences made for three-dimensional structure, perhaps not only for carriers but also for other transport proteins with similarly structured hydrophobic cores.
Work in our laboratories is supported by grants from the U.S. Public Health Service (GM24195 to P.C.M. and DK05736 to T.H.W.) and the National Science Foundation (MCB 92-20823 to P.C.M. and DCB-90-17255 to T.H.W.).
References
1. Ambudkar, S. V., V. Anantharam, and P. C. Maloney. 1990. UhpT, the sugar phosphate antiporter of Escherichia coli, functions as a monomer. J. Biol. Chem. 265:12287–12292.
2. Ambudkar, S. V., T. J. Larson, and P. C. Maloney. 1986. Reconstitution of sugar phosphate transport systems of Escherichia coli. J. Biol. Chem. 261:9083–9086.
3. Ambudkar, S. V., and P. C. Maloney. 1984. Characterization of phosphate:hexose 6-phosphate antiport in membrane vesicles of Streptococcus lactis. J. Biol. Chem. 259:12576–12585.
4. Ambudkar, S. V., L. A. Sonna, and P. C. Maloney. 1986. Variable stoichiometry of phosphate-linked anion exchange in Streptococcus lactis: implications for the mechanism of sugar phosphate transport by bacteria. Proc. Natl. Acad. Sci. USA 83:280–284.
5. Ambudkar, S. V., G. W. Zlotnick, and B. P. Rosen. 1984. Calcium efflux from Escherichia coli. Evidence for two systems. J. Biol. Chem. 259:6142–6146.
6. Anantharam, V., M. J. Allison, and P. C. Maloney. 1989. Oxalate:formate exchange: the basis for energy coupling in Oxalobacter. J. Biol. Chem. 264:7244–7250.
7. Aquila, H., T. Link, and M. Klingenberg. 1987. Solute carriers involved in energy transfer by mitochondria form a homologous protein family. FEBS Lett. 212:1–9.
8. Bassilana, M., E. Damiano-Forano, and G. Leblanc. 1985. Effect of membrane potential on the kinetic parameters of the Na+ or H+ melibiose symporter in Escherichia coli membrane vesicles. Biochem. Biophys. Res. Commun. 129:626–631.
9. Bassilana, M., T. Pourcher, and G. Leblanc. 1987. Facilitated diffusion properties of melibiose permease in Escherichia coli membrane vesicles: release of cosubstrates is rate limiting for permease cycling. J. Biol. Chem. 262:16865–16870.
10. Bennett, R. L., and M. H. Malamy. M.H. 1970. Arsenate resistant mutants of Escherichia coli and phosphate transport. Biochem. Biophys. Res. Commun. 40:496–503.
11. Beyreuther, K., B. Bieseler, R. Ehring, and B. Muller-Hill. 1981. Identification of internal residues of lactose permease of Escherichia coli by radiolabel sequencing of peptide mixtures, p. 139–148. In M. Elsinga (ed.), Methods in Protein Sequence Analysis. Humana Press, Clifton, N.J.
12. Bibi, E., G. Verner, C. Y. Chang, and H. R. Kaback. 1991. Organization and stability of a polytopic membrane protein: deletion analysis of the lactose permease of Escherichia coli. Proc. Natl. Acad. Sci. USA 88:7271–7275.
13. Blow, D. 1990. More of the catalytic triad. Nature (London) 343:694–695.
14. Blow, D. M., J. J. Birktoft, and B. S. Hartley. 1969. Role of a buried acid group in the mechanism of action of chymotrypsin. Nature (London) 221:337–340.
15. Botfield, M. C., K. Naguchi, T. Tsuchiya, and T. H. Wilson. 1992. Membrane topology of the melibiose carrier of Escherichia coli. J. Biol. Chem. 267:1818–1822.
16. Botfield, M. C., and T. H. Wilson. 1988. Mutations that simultaneously alter both sugar and cation specificity in the melibiose carrier of Escherichia coli. J. Biol. Chem. 263:12909–12915.
17. Botfield, M. C., and T. H. Wilson. 1989. Peptide-specific antibody for the melibiose carrier of Escherichia coli localizes the carboxy terminus to the cytoplasmic face of the membrane. J. Biol. Chem. 264:11649–11652.
18. Boyer, P. D. 1988. Bioenergetic coupling to protonmotive force: should we be considering hydronium ion coordination and not group protonation? Trends Biochem. Sci. 13:507.
19. Brooker, R. J. 1990. The lactose permease of Escherichia coli. Res. Microbiol. 141:309–315.
20. Brooker, R. J. 1990. Characterization of the double mutant, val-177/asn-322, of the lactose permease. J. Biol. Chem. 265:4155–4160.
21. Brooker, R. J. 1991. An analysis of lactose permease "sugar specificity" mutations which also affect the coupling between proton and lactose transport. I. Val-177 and val-177/asn-319 permeases facilitate proton uniport and sugar uniport. J. Biol. Chem. 266:4131–4138.
22. Brooker, R. J., K. Fiebig, and T. H. Wilson. 1985. Characterization of lactose carrier mutants which transport maltose. J. Biol. Chem. 260:16181–16186.
23. Brooker, R. J., and T. H. Wilson. 1985. Isolation and nucleotide sequencing of lactose carrier mutants that transport maltose. Proc. Natl. Acad. Sci. USA 82:3959–3963.
24. Brooker, R. J., and T. H. Wilson. 1986. Site-specific alteration of cysteine 176 and cysteine 234 in the lactose carrier of Escherichia coli. J. Biol. Chem. 261:11765–11769.
25. Buchel, D., B. Gronenborn, and B. Muller-Hill. 1980. Sequence of the lactose permease gene. Nature (London) 283:541–545.
26. Burant, D. F., and G. I. Bell. 1992. Mammalian facilitative glucose transporters: evidence for similar substrate recognitions sites in functionally monomeric proteins. Biochemistry 31:10414–10420.
27. Burnell, J. N., P. John, and F. R. Whatley. 1975. Phosphate transport in membrane vesicles of Paracoccus denitrificans. FEBS Lett. 58:215–218.
28. Calamia, J., and C. Manoil. 1990. Lac permease of Escherichia coli: topology and sequence elements promoting membrane insertion. Proc. Natl. Acad. Sci. USA 87:4937–4941.
29. Carrasco, N., L. M. Antes, M. S. Poonian, and H. R. Kaback. 1986. Lac permease of Escherichia coli: histidine-322 and glutamic acid-325 may be components of a charge-relay system. Biochemistry 25:4486–4488.
30. Carrasco, N., D. Herzlinger, R. Mitchell, S. DeChiara, W. Danhho, T. F. Gabriel, and H. R. Kaback. 1984. Intramolecular dislocation of the C-terminus of the lac carrier in reconstituted proteoliposomes. Proc. Natl. Acad. Sci. USA 81:4672–4676.
31. Carrasco, N., I. B. Puttner, L. M. Antes, J. A. Lee, J. D. Larigan, J. S. Lolkema, P. D. Roepe, and H. R. Kaback. 1989. Characterization of site-directed mutants in the lac permease of Escherichia coli. 2. Glutamate-325 replacements. Biochemistry 28:2533–2539.
32. Carrasco, N., P. Viitanen, H. Herzlinger, and H. R. Kaback. 1984. Monoclonal antibodies against the lac carrier protein from Escherichia coli. I. Functional studies. Biochemistry 23:3681–3687.
33. Carrasco, N., M. Tahara, L. Patel, T. Goldkorn, and R. H. Kaback. 1982. Preparation, characterization and properties of monoclonal antibodies against the lac carrier protein from Escherichia coli. Proc. Natl. Acad. Sci. USA 79:6894–6898.
34. Cohen, G. N., and J. Monod. 1957. Bacterial permeases. Bacteriol. Rev. 21:169–194.
35. Cohen, G. N., and H. V. Rickenberg. 1955. Etude directe de la fixation d’un inducteur de la β-galactosidase par les cellules d’Escherichia coli. C. R. Acad. Sci. 240:466–468.
36. Cohn, D. E., and H. R. Kaback. 1980. Mechanism of the melibiose porter in membrane vesicles of Escherichia coli. Biochemistry 19:4237–4243.
37. Collins, J. C., S. F. Permuth, and R. J. Brooker. 1989. Isolation and characterization of lactose permease mutants with an enhanced recognition of maltose and diminished recognition of cellobiose. J. Biol. Chem. 264:14698–14703.
38. Costello, M. J., J. Escaig, K. Matsushita, P. V. Viitanen, D. R. Menick, and H. R. Kaback. 1987. Purified lac permease and cytochrome o oxidase are functional as monomers. J. Biol. Chem. 262:17072–17082.
39. Crane, R. 1962. Hypothesis for mechanism of intestinal active transport of sugars. Fed. Proc. 21:891–895.
40. Damiano-Forano, E., M. Bassilana, and G. Leblanc. 1986. Sugar binding properties of the melibiose permease in Escherichia coli membrane vesicles. J. Biol. Chem. 261:6893–6899.
41. Danho, W., R. Makofske, H. Humiec, T. F. Gabriel, N. Carrisco, and H. R. Kaback. 1985. Use of site-directed polyclonal antibodies as immunotopological probes for the lac permease of Escherichia coli, p. 59–62. In Peptides, Structure and Function. Proceedings of the 9th American Peptide Symposium. Pierce Chemical Co., Rockford, Ill.
42. Davis, E. O., and P. J. F. Henderson. 1987. The cloning and DNA sequencing of the gene xylE for xylose-proton symport in E. coli. J. Biol. Chem. 262:13928–13932.
43. Deguchi, Y., I. Yamato, and Y. Anraku. 1990. Nucleotide sequence of gltS, the Na+/glutamate symport carrier of Escherichia coli B. J. Biol. Chem. 265:21704–21708.
44. Deitz, G. W. 1976. The hexose phosphate transport system of Escherichia coli. Adv. Enzymol. 44:497–509.
45. DeVoe, R. D., and P. C. Maloney. 1980. Principles of cell homeostasis, p. 3–45. In V. Mountcastle (ed.), Medical Physiology, 14th ed. The C.V. Mosby Co., St. Louis.
46. Dunten, R. L., M. Sahin-Toth, and H. R. Kaback. 1993. Role of the charge pair aspartic acid-237–lysine-358 in the lactose permease of Escherichia coli. Biochemistry 32:3139–3145.
47. Eiglmeier, K., W. Boos, and S. T. Cole. 1987. Nucleotide sequence and transcriptional startpoint of the glpT gene of Escherichia coli: extensive sequence homology of the glycerol-3-phosphate transport protein with components of the hexose 6-phosphate transport system. Mol. Microbiol. 1:251–258.
48. Essenberg, R. C., and H. L. Kornberg. 1975. Energy coupling in the uptake of hexose phosphates by Escherichia coli. J. Biol. Chem. 250:939–945.
49. Flagg, J. L., and T. H. Wilson. 1977. A protonmotive force as the source of energy for galactoside transport in energy-depleted Escherichia coli. J. Membr. Biol. 31:223–255.
50. Flagg, J. L., and T. H. Wilson. 1978. A novel type of coupling between proline and galactoside transport in Escherichia coli. Membr. Biochem. 1:61–72.
51. Foster, D. L., M. Boublik, and H. R. Kaback. 1983. Structure of the lac carrier protein of Escherichia coli. J. Biol. Chem. 258:31–34.
52. Fox, C. F., J. R. Carter, and E. P. Kennedy. 1967. Genetic control of the membrane protein component of the lactose transport system of Escherichia coli. Proc. Natl. Acad. Sci. USA 57:698–705.
53. Fox, C. F., and E. P. Kennedy. 1965. Specific labeling and partial purification of the M protein, a component of the β-galactoside transport system of Escherichia coli. Proc. Natl. Acad. Sci. USA 54:891–899.
54. Franco, P. J., and R. J. Brooker. 1991. Evidence that asparagine-322 mutant of the lactose permease transports protons and lactose with a normal stoichiometry and accumulates lactose against a concentration gradient. J. Biol. Chem. 266:6693–6699.
55. Franco, P. J., J. A. Eelkema, and R. J. Brooker. 1989. Isolation and characterization of thiodigalactoside-resident mutants of the lactose permease which possess an enhanced recognition for maltose. J. Biol. Chem. 264:15988–15992.
56. Friedberg, I. 1977. The effect of ionophores on phosphate and arsenate transport by Micrococcus lysodeikticus. FEBS Lett. 81:264–266.
57. Friedrich, M. J., and R. J. Kadner. 1987. Nucleotide sequence of the uhp region of Escherichia coli. J. Bacteriol. 169:3556–3563.
58. Goldkorn, T., G. Rimon, and H. R. Kaback. 1983. Topology of the lac carrier protein in the membrane of Escherichia coli. Proc. Natl. Acad. Sci. USA 80:3322–3326.
59. Goldrick, D., G. Q. Yu, S. Q. Jiang, and J. S. Hong. 1988. Nucleotide sequence and transcription start point of the phosphoglycerate transporter gene of Salmonella typhimurium. J. Bacteriol. 179:3421–3426.
60. Gott, P., and W. Boos. 1988. The transmembrane topology of the sn-glycerol-3-phosphate permease of Escherichia coli analyzed by phoA and lacZ protein fusions. Mol. Microbiol. 2:655–663.
61. Gram, C. D., and R. J. Brooker. 1992. An analysis of the side chain requirement at position 177 within the lactose permease which confers the ability to recognize maltose. J. Biol. Chem. 267:3841–3846.
62. Griffith, J. K., M. E. Baker, D. A. Rouch, M. G. P. Page, R. A. Skurray, I. T. Paulsen, K. F. Chater, S. A. Baldwin, and P. J. F. Henderson. 1992. Membrane transport proteins: implications of sequence comparisons. Curr. Opin. Cell Biol. 4:684–695.
63. Hama, H., and T. H. Wilson. 1992. Primary structure and characteristics of the melibiose carrier of Klebsiella pneumoniae. J. Biol. Chem. 267:18371–18376.
64. Hama, H., and T. H. Wilson. 1993. Cation-coupling in chimeric melibiose carriers derived from Escherichia coli and Klebsiella pneumoniae. The amino-terminal portion is crucial for Na+ recognition in melibiose transport. J. Biol. Chem. 268:10060–10065.
65. Hama, H., and T. H. Wilson. 1994. Replacement of alanine-58 by asparagine enables the melibiose carrier of Klebsiella pneumoniae to couple sugar transport to Na+. J. Biol. Chem. 269:1063–1067.
66. Harold, F. M. 1977. Membranes and energy transduction in bacteria. Curr. Top. Bioenerg. 6:83–149.
67. Harold, F. M., R. L. Harold, and A. Abrams. 1965. A mutant of Streptococcus faecalis defective in phosphate transport. J. Biol. Chem. 240:3145–3153.
68. Hayashi, S.-I., J. P. Koch, and E. C. C. Lin. 1964. Active transport of L-α-glycerophosphate in Escherichia coli. J. Biol. Chem. 239:3098–3105.
69. Heller, K. B., E. C. C. Lin, and T. H. Wilson. 1980. Substrate specificity and transport properties of the glycerol facilitator of Escherichia coli. J. Bacteriol. 144:274–278.
70. Henderson, P. J. F. 1990. Proton-linked sugar transport systems in bacteria. J. Bioenerg. Biomembr. 22:525–569.
71. Henderson, P. J. F. 1993. The 12-transmembrane helix transporters. Curr. Opin. Cell Biol. 5:708–721.
72. Henderson, P. J. F., S. A. Baldwin, M. T. Cairns, B. M. Charalambous, H. C. Dent, F. Gunn, W. J. Liang, V. A. Lucas, G. E. Martin, T. P. McDonald, B. J. McKeown, J. A. R. Muiry, K. R. Petero, P. E. Roberts, K. R. Shatwell, G. Smith, and C. G. Tate. 1992. Sugar-cation symport systems in bacteria. Int. Rev. Cytol. 137A:149–208.
73. Herbert, D. N., and A. Carruthers. 1992. Glucose transporter oligomeric structure determines transporter function. Reversible redox-dependent interconversions of tetrameric and dimeric GLUT1. J. Biol. Chem. 267:23829–23838.
74. Herzlinger, D., N. Carrasco, and H. R. Kaback. 1985. Functional and immunochemical characterization of a mutant of Escherichia coli energy-coupled for lactose transport. Biochemistry 24:221–229.
75. Herzlinger, D., P. Viitanen, N. Carrasco, and H. R. Kaback. 1984. Monoclonal antibodies against the lac carrier protein from Escherichia coli. II. Binding studies with membrane vesicles and proteoliposomes reconstituted with purified lac carrier protein. Biochemistry 23:3688–3693.
76. Hirata, H., K. Altendorf, and F. M. Harold. 1973. Role of an electrical potential in the coupling of metabolic energy to active transport by membrane vesicles of Escherichia coli. Proc. Natl. Acad. Sci. USA 70:1804–1808.
77. Island, M. D., B. Y. Wei, and R. J. Kadner. 1992. Structure and function of the uhp genes for the sugar phosphate transport system in Escherichia coli and Salmonella typhimurium. J. Bacteriol. 174:2754–2762.
78. Ivey, D. M., A. A. Guffanti, J. Zemsky, E. Pinner, R. Karpel, E. Padan, S. Schuldiner, and T. A. Krulwich. 1993. Cloning and characterization of a putative Ca2+/H+ antiporter gene from Escherichia coli upon functional complementation of Na+/H+ antiporter-deficient strains by the overexpressed gene. J. Biol. Chem. 268:11296–11303.
79. Iwaarden, P. R., J. C. Pastore, W. N. Konings, and H. R. Kaback. 1991. Construction of a functional lactose permease devoid of cysteine residues. Biochemistry 30:9595–9600.
80. Jank, B., B. Habermann, R. J. Scheweyen, and R. J. Link. 1992. PMP47, a peroxisomal homologue of mitochondrial solute carrier proteins. Trends Biochem. Sci. 18:427–428.
81. Jones, T. H. D., and E. P. Kennedy. 1969. Characterization of the membrane protein component of the lactose transport system of Escherichia coli. J. Biol. Chem. 244:5981–5987.
82. Jung, K., H. Jung, and H. R. Kaback. 1994. Dynamics of lactose permease of Escherichia coli determined by site-directed fluorescence labeling. Biochemistry 33:3980–3985.
83. Jung, K., H. Jung, J. Wu, G. G. Prive and H. R. Kaback. 1993. Use of site-directed fluorescence labeling to study proximity relationships in the lactose permease of Escherichia coli. Biochemistry 32:12273–12278.
84. Kaback, H. R. 1987. Use of site-directed mutagenesis to study the mechanism of a membrane transport protein. Biochemistry 26:2071–2076.
85. Kaback, H. R. 1988. Site-directed mutagenesis and ion-gradient driven active transport: on the path of the proton. Annu. Rev. Physiol. 50:243–256.
86. Kaczorowski, G. J., G. LeBlanc, and H. R. Kaback. 1980. Specific labeling of the lac carrier protein in membrane vesicles of Escherichia coli by a photoaffinity reagent. Proc. Natl. Acad. Sci. USA 77:6319–6323.
87. Kadner, R. J., C. A. Webber, and M. D. Island. 1993. The family of organophosphate transport proteins includes a transmembrane regulatory protein. J. Bioenerg. Biomembr. 25:637–645.
88. Kahan, F. M., J. S. Kahan, P. J. Cassidy, and H. Kropp. 1974. The mechanism of action of fosfomycin (phosphonomycin). Ann. N.Y. Acad. Sci. 235:364–386.
89. Kashket, E. R., and T. H. Wilson. 1973. Proton-coupled accumulation of galactoside in Streptococcus lactis 7962. Proc. Natl. Acad. Sci. USA 70:2866–2869.
90. Kawakami, T., Y. Akizawa, T. Ishikawa, T. Shimamoto, M. Tsuda, and T. Tsuchiya. 1988. Amino acid substitutions and alteration in cation specificity in the melibiose carrier of Escherichia coli. J. Biol. Chem. 263:14276–14280.
91. Kennedy, E. P. 1970. The lactose permease system of Escherichia coli, p. 49–92. In J. R. Beckwith and D. Zipser (ed.), The Lactose Operon. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
92. King, S. C., C. L. Hansen, and T. H. Wilson. 1991. The interaction between aspartic acid 237 and lysine 358 in the lactose carrier of Escherichia coli. Biochim. Biophys. Acta 1062:177–186.
93. King, S. C., and T. H. Wilson. 1989. Galactoside-dependent proton transport by mutants of the Escherichia coli lactose carrier: substitution of tyrosine for histidine-322 and of leucine for serine-306. Biochim. Biophys. Acta 982:253–264.
94. King, S. C., and T. H. Wilson. 1989. Galactoside-dependent proton transport by mutants of the Escherichia coli lactose carrier: replacement of histidine-322 by tyrosine or phenylalanine. J. Biol. Chem. 264:7390–7394.
95. King, S. C., and T. H. Wilson. 1990. Identification of valine 177 as a mutation altering specificity for transport of sugars by the Escherichia coli lactose carrier: enhanced specificity for sucrose and maltose. J. Biol. Chem. 265:9638–9644.
96. Konings, W. N., and G. T. Robillard. 1982. Physical mechanism for regulation of proton solute symport in Escherichia coli. Proc. Natl. Acad. Sci. USA 79:5480–5484.
97. Kramer, R. 1994. Functional principles of solute transport systems: concepts and perspectives. Biochim. Biophys. Acta 1185:1–34.
98. Kuan, J., and M. H. Saier, Jr. 1993. The mitochondrial carrier family of transport proteins: structural, functional and evolutionary relationships. Crit. Rev. Biochem. Mol. Biol. 28:209–233.
99. Kyte, J., and R. F. Doolittle. 1982. A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157:105–132.
100. Lancaster J. R., Jr., and P. C. Hinkle. 1977. Studies of the β-galactoside transporter in inverted membrane vesicles of Escherichia coli. I. Symmetrical facilitated diffusion and proton gradient-coupled transport. J. Biol. Chem. 252:7657–7661.
101. Lee, J. I., P. P. Hwang, C. Hanson, and T. H. Wilson. 1992. Possible salt bridges between transmembrane α-helices of the lactose carrier of Escherichia coli. J. Biol. Chem. 267:20758–20764.
102. Lee, J. I., P. P. Hwang, and T. H. Wilson. 1993. Lysine 319 interacts with both glutamic acid 269 and aspartic acid 240 in the lactose carrier of Escherichia coli. J. Biol. Chem. 268:20007–20015.
103. Leong-Morgenthaler, P., M. C. Zwahlen, and H. Hottinger. 1991. Lactose metabolism in Lactobacillus bulgaricus: analysis of the primary structure and expression of the genes involved. J. Bacteriol. 173:1951–1957.
104. Lindenthal, S., and D. Schubert. 1991. Monomeric erythrocyte band 3 protein transports anions. Proc. Natl. Acad. Sci. USA 88:6540–6544.
105. Linial, M. 1993. Vesicular transporters join the major facilitator superfamily (MFS). Trends Biochem. Sci. 18:248–249.
106. Maiden, M. C. J., E. O. Davis, S. A. Baldwin, D. C. M. Moore, and P. J. F. Henderson. 1987. Mammalian and bacterial sugar transport proteins are homologous. Nature (London) 325:641–643.
107. Maiden, M. C. J., M. C. Jones-Mortimer, and P. J. F. Henderson. 1988. The cloning, DNA sequence and overexpression of the gene araE coding for arabinose-proton symport in Escherichia coli K12. J. Biol. Chem. 263:8003–8010.
108. Maloney, P. C. 1987. Coupling to an energized membrane: role of ion-motive gradients in the transduction of metabolic energy, p. 222–243. In F. C. Neidhardt, J. L. Ingraham, K. B. Low, B. Magasanik, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology, vol. 1. American Society for Microbiology, Washington, D.C.
109. Maloney, P. C. 1990. A consensus structure for membrane transport. Res. Microbiol. 141:374–383.
110. Maloney, P. C. 1994. Bacterial transporters. Curr. Opin. Cell Biol. 6:571–582.
111. Maloney, P. C., S. V. Ambudkar, V. Anantharam, L. A. Sonna, and A. Varadhachary. 1990. Anion-exchange mechanisms in bacteria. Microbiol. Rev. 54:1–17.
112. Maloney, P. C., S. V. Ambudkar, J. Thomas, and L. Schiller. 1984. Phosphate:hexose 6-phosphate antiport in Streptococcus lactis. J. Bacteriol. 158:238–245.
113. Maloney, P. C., and T. H. Wilson. 1993. The evolution of membrane carriers, p. 147–160. In O. Reuss, J. M. Russell, Jr., and M. L. Jennings (ed.), Molecular Biology and Function of Carrier Proteins. Rockefeller University Press, New York.
114. Marger, M. D., and M. H. Saier, Jr. 1993. A major superfamily of transmembrane facilitators that catalyse uniport, symport and antiport. Trends Biochem. Sci. 18:13–20.
115. Markgraf, M., H. Bocklage, and B. Muller-Hill. 1985. A change of threonine 266 to isoleucine in the lac permease of Escherichia coli diminishes the transport of lactose and increases the transport of maltose. Mol. Gen. Genet. 198:473–475.
116. Matzke, E. A., L. J. Stephenson, and R. J. Brooker. 1992. Functional role of arginine 302 within the lactose permease of Escherichia coli. J. Biol. Chem. 267:19095–19100.
117. Maurel, C., J. Reizer, J. I. Schroeder, M. J. Chrispeels, and M. H. Saier, Jr. 1994. Functional characterization of the Escherichia coli glycerol facilitator, GlpF, in Xenopus oocytes. J. Biol. Chem. 269:11869–11872.
118. Menezes, M. E., P. D. Roepe, and H. R. Kaback. 1990. Design of a membrane transport protein for fluorescence spectroscopy. Proc. Natl. Acad. Sci. USA 87:1638–1642.
119. Meng, S.-Y., and G. N. Bennet. 1992. Nucleotide sequence of the Escherichia coli cad operon: a system for neutralization of low extracellular pH. J. Bacteriol. 174:2659–2669.
120. Menick, D. R., N. Carrasco, L. Antes, L. Patel, and H. R. Kaback. 1987. Lac permease of Escherichia coli: arginine-302 as a component of the postulated proton relay. Biochemistry 26:6638–6644.
121. Menick, D. R., J. A. Lee, R. J. Brooker, T. H. Wilson, and H. R. Kaback. 1987. Role of cysteine residues in the Lac permease of Escherichia coli. Biochemistry 26:1132–1136.
122. Menick, D. R., H. K. Sarkar, M. S. Poonian, and H. R. Kaback. 1985. Cys154 is important for lac permease activity in Escherichia coli. Biochem. Biophys. Res. Commun. 132:162–170.
123. Mitchell, P. 1954. Transport of phosphate across the surface of Micrococcus pyogenes: nature of the cell ‘inorganic phosphate’. J. Gen. Microbiol. 9:273–287.
124. Mitchell, P. 1963. Molecule, group and electron translocation through natural membranes. Biochem. Soc. Symp. 22:142–168.
125. Mitchell, P. 1967. Translocations through natural membranes. Adv. Enzymol. 29:33–87.
126. Mitchell, P., and J. Moyle. 1953. Paths of phosphate transfer in Micrococcus pyogenes: phosphate turnover in nucleic acid and other fractions. J. Gen. Microbiol. 9:257–272.
127. Mizushima, K., S. Awakihara, M. Kuroda, and T. Ishikawa. 1992. Cloning and sequencing of the melB gene encoding the melibiose permease of Salmonella typhimurium LT2. Mol. Gen. Genet. 234:74–80.
128. Molenaar D., J. S. Bosscher, B. TenBrink, A. J. M. Driessen, and W. N. Konings. 1993. Generation of a proton motive force by histidine decarboxylation and electrogenic histidine/histamine antiport in Lactobacillus buchneri. J. Bacteriol. 175:2864–2870.
129. Motojima, K., I. Yamato, Y. Anraku, A. Nishimura, and Y. Hirota. 1979. Amplification and characterization of the proline transport carrier of Escherichia coli K-12 by using proT + hybrid plasmids. Proc. Natl. Acad. Sci. USA 76:6255–6259.
130. Nelson, D. R., J. E. Lawson, M. Klingenberg, and M. G. Douglas. 1993. Site- directed mutagenesis of the yeast mitochondrial ADP/ATP translocator. Six arginines and one lysine are essential. J. Mol. Biol. 230:1159–1170.
131. Neuhaus, J. M., J. Soppa, J. K. Wright, I. Reide, H. Blocker, R. Frank, and P. Overath. 1985. Properties of a mutant lactose carrier of Escherichia coli with a Cys148→ Ser148 substitution. FEBS Lett. 185:83–88.
132. Newman, M. J., D. Foster, T. H. Wilson, and H. R. Kabak. 1981. Purification and reconstitution of functional lactose carrier from Escherichia coli. J. Biol. Chem. 256:11804–11808.
133. Newman, M. J., and T. H. Wilson. 1980. Solubilization and reconstitution of the lactose transport system from Escherichia coli . J. Biol. Chem. 255:10583–10586.
134. Niiya, S., K. Yamasaki, T. H. Wilson, and T. Tsuchiya. 1982. Altered cation coupling to melibiose transport in mutants of Escherichia coli. J. Biol. Chem. 257:8902–8906.
135. Nilsson R. P., L. Beijer, and B. Rutberg. 1994. The glpT and glpQ genes of the glycerol regulon in Bacillus subtilis. Microbiology 140:723–730.
136. Olsen, S. G., K. M. Greene, and R. J. Brooker. 1993. Lactose permease mutants which transport (malto-)oligosaccarides. J. Bacteriol. 175:6269–6275.
137. Padan, E., H. R. Sarkar, P. V. Viitanen, M. S. Poonian, and H. R. Kaback. 1985. Site-specific mutagenesis of histidine residues in the lac permease of Escherichia coli. Proc. Natl. Acad. Sci. USA 82:6765–6768.
138. Padan, E., and S. Schuldiner. 1994. Molecular physiology of Na+/H+ antiporters, key transporters in circulation of Na+ and H+ in cells. Biochim. Biophys. Acta 1185:129–151.
139. Page, M. G. P., and J. P. Rosenbusch. 1988. Topography of lactose permease from Escherichia coli. J. Biol. Chem. 263:15906–15914.
140. Pardee, A. B. 1957. An inducible mechanism for accumulation of melibiose in Escherichia coli. J. Bacteriol. 73:376–385.
141. Pogell, B. M., B. R. Maity, S. Frumkin, and S. Shapiro. 1966. Induction of an active transport system for glucose 6-phosphate in Escherichia coli. Arch. Biochem. Biophys. 116:406–415.
142. Poolman, B., and W. N. Konings. 1993. Secondary solute transport in bacteria. Biochim. Biophys. Acta 1183:5–39.
143. Poolman, B., D. Molenaar, E. J. Smid, T. Ubbink, T. Abee, P. P. Renault, and W. N. Konings. 1991. Malolactic fermentation: electrogenic malate uptake and malate/lactate antiport generate metabolic energy. J. Bacteriol. 173:6030–6037.
144. Poolman, B., T. J. Royer, S. E. Mainzer, and B. F. Schmidt. 1989. Lactose transport system of Streptococcus thermophilus. J. Bacteriol. 71:244–253.
145. Pourcher, T., M. Bassilana, H. K. Sarkar, H. R. Kaback, and G. Leblanc. 1990. Na+/melibiose symport mechanism of Escherichia coli: kinetic and molecular properties. Phil. Trans. R. Soc. London Ser. B 326:411–423.
146. Pourcher, T., M. Bassilana, H. K. Sarkar, H. R. Kaback, and G. Leblanc. 1990. Melibiose permease and α-galactosidase of Escherichia coli: identification by selective labeling using the T7 RNA polymerase/promoter expression system. Biochemistry 29:690–696.
147. Pourcher, T., M. Bassilana, H. K. Sarkar, H. R. Kaback, and G. Leblanc. 1992. The melibiose permease of Escherichia coli: mutations of His94 alter insertion and stability into the membrane rather than catalytic activity. Biochemistry 31:5225–5231.
148. Pourcher, T., H. K. Sarkar, M. Bassilana, H. R. Kaback, and G. Leblanc. 1990. Histidine-94 is the only important histidine residue in the melibiose permease of Escherichia coli. Proc. Natl. Acad. Sci. USA 87:468–472.
149. Pourcher, T., M. L. Zani, and G. Leblanc. 1993. Mutagenesis of acidic residues in putative membrane-spanning segments of the melibiose permease of Escherichia coli. I. Effect on Na+-dependent transport and binding properties. J. Biol. Chem. 268:3209–3215.
150. Prestige, L. S., and A. B. Pardee. 1965. A second permease for methyl-thio-β-d-galactoside in Escherichia coli. Biochim. Biophys. Acta 100:591–593.
151. Puttner, I. B., H. K. Sarkar, E. Padan, J. S. Lolkema, and H. R. Kaback. 1989. Characterization of site-directed mutants in the lac permease of Escherichia coli. 1. Replacement of histidine residues. Biochemistry 28:2525–2533.
152. Puttner, I. B., H. K. Sarkar, M. S. Poonian, and H. R. Kaback. 1986. Lac permease of Escherichia coli : histidine-205 and histidine-322 play different roles in lactose/H+ symport. Biochemistry 25:4483–4485.
153. Ramos, S., and H. R. Kaback. 1977. The relationship between the electrochemical proton gradient and active transport in Escherichia coli membrane vesicles. Biochemistry 16:854–859.
154. Ramos, S., and H. R. Kaback. 1977. The electrochemical proton gradient in Escherichia coli membrane vesicles. Biochemistry 16:848–854.
155. Reithmeier, R. A. F. 1993. The erythrocyte anion transorter (band 3). Curr. Opin. Struct. Biol. 3:515–523.
156. Reizer, J., A. Reizer, and M. H. Saier, Jr. 1993. The MIP family of integral membrane channel proteins: sequence comparisons, evolutionary relationships, reconstructed pathway of evolution, and proposed functional differentiation of the two repeated halves of the proteins. Crit. Rev. Biochem. Mol. Biol. 28:235–257.
157. Reizer, J., A. Reizer, and M. H. Saier, Jr. 1994. A functional superfamily of sodium/solute symporters. Biochim. Biophys. Acta 1197:133–166.
158. Rickenberg, H. V., G. N. Cohen, G. Butlin, and J. Monod. 1956. La galactoside-permease d’Escherichia coli. Ann. Inst. Pasteur 91:829–857.
159. Roepe, P. D., T. G. Consler, M. E. Menzes, and H. R. Kaback. 1990. The lac permease of Escherichia coli: site-directed mutagenesis studies on the mechanism of β-galactoside/H+ symport. Res. Microbiol. 141:290–308.
160. Roepe, P. D., and H. R. Kaback. 1989. Site-directed mutagenesis of tyrosine residues in the lac permease of Escherichia coli. Biochemistry 28:6127–6132.
161. Rosen, B. P., and E. R. Kashket. 1978. Energetics of active transport, p. 559–620. In B. P. Rosen (ed.), Bacterial Transport. Marcel Dekker, Inc., New York.
162. Rosen, B. P., and J. S. McClees. 1974. Active transport of calcium in inverted membrane vesicles of E. coli. Proc. Natl. Acad. Sci. USA 71:5042–5046.
163. Rosenberg, H. 1987. Phosphate transport in prokaryotes p. 559–620. In B. P. Rosen and S. Silver (ed.), Ion Transport in Prokaryotes. Academic Press, Inc., New York.
164. Rosenberg, H., R. G. Gerdes, and K. Chegwidden. 1977. Two systems for the uptake of phosphate by Escherichia coli. J. Bacteriol. 131:505–511.
165. Rosenberg, H., R. G. Gerdes, and F. M. Harold. 1979. Energy coupling to the transport of inorganic phosphate in Escherichia coli. J. Bacteriol. 131:505–511.
166. Rotman, M. B. 1958. Regulation of enzymatic activity in the intact cell: the β-galactosidase of Escherichia coli. J. Bacteriol. 76:1–14.
167. Ruan, Z.-S., V. Anantharam, I. T. Crawford, S. V. Ambudkar, S. Y. Rhee, M. J. Allison, and P. C. Maloney. 1992. Identification, purification and reconstitution of OxlT, the oxalate:formate antiport protein of Oxalobacter formigenes. J. Biol. Chem. 267:10537–10543.
168. Rubin, R. R., S. B. Levy, R. L. Heinrikson, and F. J. Kezdy. 1990. Gene duplication in the evolution of the two complementing domains of gram-negative bacterial tetracycline efflux proteins. Gene 87:7–13.
169. Sahin-Toth, M., R. L. Dunsten, A. Gonzalez, and H. R. Kaback. 1992. Functional interactions between putative intramembrane charged residues in the lactose permease of Escherichia coli. Proc. Natl. Acad. Sci. USA 89:10541–10551.
170. Sahin-Toth, M., R. L. Dunten, and H. R. Kaback. 1993. Molecular Biology and Function of Carrier Proteins, p. 2–8. Rockefeller University Press.
171. Saier, M. H., Jr., D. L. Wentzel, B. U. Feucht, and J. J. Judice. 1975. A transport system for phosphoenolpyruvate, 2-phosphoglycerate and 3-phosphoglycerate in Salmonella typhimurium. J. Biol. Chem. 250:5089–5096.
172. Sanno, Y., T. H. Wilson, and E. C. C. Lin. 1968. Control of permeation to glycerol in cells of Escherichia coli. Biochem. Biophys. Res. Commun. 32:344–349.
173. Sarero, J. P., and A. J. Pittard. 1995. Membrane topology analysis of Escherichia coli K-12 MTR permease by alkaline phosphatase and β-galactosidase fusions. J. Bacteriol. 177:297–306.
174. Sarkar, H. K., D. R. Menick, P. V. Viitanen, M. S. Poonian, and H. R. Kaback. 1986. Site-specific mutagenesis of cysteine 148 to serine in the lac permease of Escherichia coli. J. Biol. Chem. 261:8914–8918.
175. Schuldiner, S., and E. Padan. 1992. Na+/H+ antiporters in Escherichia coli, p. 25–51. In E. P. Bakker (ed.), Alkali Cation Transport Systems in Prokaryotes. CRC Press, Inc., Boca Raton, Fla.
176. Seckler, R., T. Moroy, J. K. Wright, and P. Overath. 1986. Anti-peptide antibodies and proteases as structural probes for the lactose/H+ transporter of Escherichia coli: a loop around amino acid residue 130 faces the cytoplasmic side of the membrane. Biochemistry 25:2403–2409.
177. Seckler, R., and J. K. Wright. 1984. Sidedness of native membrane vesicles of Escherichia coli and orientation of the reconstituted lactose:H+ carrier. Eur. J. Biochem. 142:269–279.
178. Seckler, R., J. K. Wright, and P. Overath. 1983. Peptide-specific antibody locates the COOH terminus of the lactose carrier of Escherichia coli on the cytoplasmic side of the plasma membrane. J. Biol. Chem. 258:10817–10820.
179. Shiota, S., Y. Yamane, M. Futai, and T. Tsuchiya. 1985. Escherichia coli mutants possessing an Li+-resistant melibiose carrier. J. Bacteriol. 162:106–109.
180. Shiota, S., H. Yazyu, and T. Tsuchiya. 1984. Escherichia coli mutants with altered cation recognition by the melibiose carrier. J. Bacteriol. 160:445–447.
181. Shuman, H. A., and J. Beckwith. 1979. Escherichia coli K-12 mutants that allow transport of maltose via the β-galactoside transport system. J. Bacteriol. 137:365–373.
182. Sonna, L. A., S. V. Ambudkar, and P. C. Maloney. 1988. The mechanism of glucose 6-phosphate transport by Escherichia coli. J. Biol. Chem. 263:6625–6630.
183. Sonna, L. A., and P. C. Maloney. 1988. Identification and functional reconstitution of phosphate:sugar phosphate antiport of Staphylococcus aureus. J. Membr. Biol. 101:267–274.
184. Stein, W. D. 1986. Transport and Diffusion Across Cell Membranes. Academic Press, Inc., Orlando, Fla.
185. Stewart, L. M. D., and I. R. Booth. 1983. Na+ involvement in proline transport in Escherichia coli. FEMS Microbiol. Lett. 19:161–164.
186. Stochaj, U., B. Bieseler, and R. Ehring. 1986. Limited proteolysis of lactose permease from Escherichia coli. Eur. J. Biochem. 158:423–428.
187. Stochaj, U., and R. Ehring. 1987. The N-terminal region of Escherichia coli lactose permease mediates membrane contact of the nascent polypeptide chain. Eur. J. Biochem. 163:653–658.
188. Stochaj, U., H. J. Fritz, C. Heibach, M. Markgraf, A. von Schaewen, U. Sonnewald, and R. Ehring. 1990. Truncated forms of Escherichia coli lactose permease: models for study of biosynthesis and membrane insertion. J. Bacteriol. 170:2639–2645.
189. Sweet, G., C. Gandor, R. Voegele, N. Wittekindt, J. Beuerle, V. Truniger, E. C. Lin, and W. Boos. 1990. Glycerol facilitator of Escherichia coli: cloning of glpF and identification of the GlpF product. J. Bacteriol. 172:424–430.
190. Taglicht, D., E. Padan, and S. Schuldiner. 1991. Overproduction and purification of a functional Na+/H+ antiporter coded by nhaA (ant) from Escherichia coli. J. Biol. Chem. 266:11289–11294.
191. Teather, R. M., J. Bramhall, I. Riede, J. K. Wright, M. Furst, G. Aichele, V. Wilhelm, and P. Overath. 1980. Lactose carrier protein of Escherichia coli: structure and expression of plasmids carrying the y gene of the lac operon. Eur. J. Biochem. 108:223–231.
192. Trumble, W. R., P. V. Viitanen, H. K. Sarkar, M. S. Poonian, and H. R. Kaback. 1984. Site-directed mutagenesis of Cys148 in the lac carrier protein of Escherichia coli. Biochem. Biophys. Res. Commun. 119:860–867.
193. Tsuchiya, T., J. Lopilato, and T. H. Wilson. 1978. Effect of lithium ion on melibiose transport in Escherichia coli. J. Membr. Biol. 42:45–59.
194. Tsuchiya, T., M. Oho, and S. Shiota-Niiya. 1983. Lithium ion-sugar cotransport via the melibiose transport system in Escherichia coli. Measurement of Li+ transport and specificity. J. Biol. Chem. 258:12765–12767.
195. Tsuchiya, T., J. Raven, and T. H. Wilson. 1977. Co-transport of Na+ and methyl-β-D-thiogalactopyranoside mediated by the melibiose transport system of Escherichia coli. Biochem. Biophys. Res. Commun. 76:26–31.
196. Tsuchiya, T., D. M. Wilson, and T. H. Wilson. 1985. Melibiose-cation cotransport system of Escherichia coli. Ann. N.Y. Acad. Sci. 456:342–349.
197. van Veen, H. W., T. Abee, G. J. J. Kortstee, W. N. Konings, and A. J. B. Zehnder. 1994. Translocation of metal phosphate via the phosphate inorganic transport system of Escherichia coli. Biochemistry 33:1766–1770.
198. Varadhachary, A., and P. C. Maloney. 1991. Reconstitution of the phosphoglycerate transport protein of Salmonella typhimurim. J. Biol. Chem. 266:130–135.
199. Viitanen, P., M. L. Garcia, and H. R. Kaback. 1984. Purified reconstituted lac carrier protein from Escherichia coli is fully functional. Proc. Natl. Acad. Sci. USA 89:1629–1633.
200. Viitanen, P. V., D. R. Menick, H. K. Sarkar, W. R. Trumble, and H. R. Kaback. 1985. Site-directed mutagenesis of cysteine-148 in the lac permease of Escherichia coli: effect on transport, binding, and sulfhydryl inactivation. Biochemistry 24:7628–7635.
201. Vogel, H., J. K. Wright, and F. Jahnig. 1985. The structure of the lactose permease derived from Raman spectroscopy and prediction methods. EMBO J. 4:3625–3631.
202. von Heijne, G. 1989. Control of topology and mode of assembly of a polytopic membrane protein by positively charged residues. Nature (London) 341:456–458.
203. Wagner, R., E. C. Apley, G. Gross, and U. I. Flugge. 1989. The rotational diffusion of chloroplast phosphate translocator and of lipid molecules in bilayer membranes. Eur. J. Biochem. 182:165–173.
204. Walmsley, A. R., K. R. Petro, and P. J. F. Henderson. 1993. Equilibrium and transient kinetic studies of the binding of cytochalasin B to the L-arabinose-H+ symport protein of Escherichia coli. Determination of the sugar binding specificity of the l-arabinose-H+ symporter. Eur. J. Biochem. 215:43–45.
205. West, I. C. 1970. Lactose transport coupled to proton movements in Escherichia coli. Biochem. Biophys. Res. Commun. 41:655–661.
206. West, I. C., and P. Mitchell. 1972. Proton-coupled β-galactoside translocation in non-metabolizing Escherichia coli. J. Bioenerg. 3:445–462.
207. Widdas, W. F. 1952. Inability of diffusion to account for placental glucose transfer in the sheep and consideration of the kinetics of a possible carrier transfer. J. Physiol. 118:23–39.
208. Widdas, W. F. 1988. Old and new concepts of the membrane transport for glucose in cells. Biochim. Biophys. Acta 947:385–404.
209. Willsky, G. R., and M. H. Malamy. 1980. Characterization of two genetically separable inorganic phosphate transport systems in Escherichia coli. J. Bacteriol. 144:366–374.
210. Wilson, D. M., and T. H. Wilson. 1987. Cation specificity for sugar substrates of the melibiose carrier in Escherichia coli. Biochim. Biophys. Acta 904:191–200.
211. Wilson, D. M., and T. H. Wilson. 1992. Asp-51 and Asp-120 are important for the transport function of the Escherichia coli melibiose carrier. J. Bacteriol. 174:3083–3086.
212. Winkler, H. H. 1966. A hexose-phosphate transport system in Escherichia coli. Biochim. Biophys. Acta 117:231–240.
213. Winkler, H. H. 1973. Energy coupling of the hexose phosphate transport system in Escherichia coli. J. Bacteriol. 116:203–209.
214. Winkler, H. H. 1973. Distribution of an inducible hexose-phosphate transport system among various bacteria. J. Bacteriol. 116:1079–1081.
215. Wood, J. M., D. Zadworney, E. Lohmeiner, and J. H. Weiner. 1979. Hybrid plasmids complement a putP mutation in Escherichia coli K12. Can. J. Biochem. 57:1328–1330.
216. Yamaguchi, A., Y. Iwasaki-Ohba, N. Ono, M. Kaneko-Ohdera, and T. Sawai. 1991. Stoichiometry of metal-tetracycline/H+ antiport mediated by transposon Tn10-encoded tetracycline resistance protein in Escherichia coli. FEBS Lett. 282:415–418.
217. Yamaguchi, A., N. Ono, T. Akasaka, T. Noumi, and T. Sawai. 1990. Metal-tetracycline/H+ antiporter of Escherichia coli encoded by a transposon, Tn10: the role of the conserved dipeptide, Ser65-Asp66, in tetracycline transport. J. Biol. Chem. 265:15525–15530.
218. Yamaguchi, A., S. Yuichi, and T. Sawa. 1992. Metal-tetracycline/H+ antiporter of Escherichia coli encoded by transposon Tn10: the role of a conserved sequence motif, GXXXXRXGRR, in a putative cytoplasmic loop between helices 2 and 3. J. Biochem. 267:19155–19162.
219. Yamato, I., and Y. Anraku. 1988. Site-specific alteration of cysteine 281, cysteine 344 and cysteine 349 in the proline carrier of Escherichia coli. J. Biol. Chem. 263:16055–16057.
220. Yan, R.-T., and P. C. Maloney. 1993. Identification of a residue in the translocation pathway of a membrane carrier. Cell 75:37–44.
220a. Yan, R.-T., and P. C. Maloney. 1995. Residues in the pathway through a membrane transporter. Proc. Natl. Acad. Sci. USA 92:5973–5976.
221. Yazyu, H., S. Shiota, M. Futai, and T. Tsuchiya. 1985. Alteration in cation specificity of the melibiose transport carrier of Escherichia coli due to replacement of proline 122 with serine. J. Bacteriol. 162:933–937.
222. Yazyu, H., S. Shiota-Niiya, T. Shimamoto, H. Kanazawa, M. Futai, and T. Tsuchiya. 1984. Nucleotide sequence of the melB gene and characteristics of deduced amino acid sequence of the melibiose carrier in Escherichia coli. J. Biol. Chem. 259:4320–4326.
223. Zani, M. L., T. Pourcher, and G. Leblanc. 1993. Mutagenesis of acidic residues in putative membrane-spanning segments of the melibiose permease of Escherichia coli. II. Effect on cationic selectivity and coupling properties. J. Biol. Chem. 268:3216–3221.
Chapter74
